extremely serious – even life-threatening.
Eyewitness testimony in criminal trials can be, to put it simply, unreliable. As Buckhout
notes:
Research and courtroom experience provide ample
evidence that an eyewitness to a crime is being asked to be
something and do something that a normal human being
was not created to be or do. Human perception is sloppy
and uneven, albeit remarkably effective in serving our need
to create structure out of experience. In an investigation or
in court, however, a witness is often asked to play the role
of a kind of tape recorder on whose tape the events of a
crime have left an impression…. Both sides, and usually
the witness, too, succumb to the fallacy that everything was
recorded and can be played back without questioning
(1974, 23).
Various factors contributing to the unreliability of eyewitness identification have been
explored by a number of researchers (see Buckhout, Figueroa, and Hoff, 1975; Clifford and
42
Scott, 1978; Loftus, 1979; Shapiro and Penrod 1986). Considerable research advancements have
been made in the past 25 years concerning such topics as: how jurors evaluate eyewitness
testimony (Cutler, Penrod, and Stuve 1988; Wells 1984), the reconstructive aspects of
eyewitness memories (Loftus 1979), the development of false autobiographical memories
(Linsay and Read 1995; Loftus and Ketcham 1994), the effectiveness of legal safeguards
(Stinson, Devenport, Cutler, and Kravitz 1996), individual differences in eyewitness abilities
(Hosch 1994), techniques for interviewing eyewitnesses (Fisher 1995), the abilities of children
relative to adults to recall witnessed events (Ceci, Toglia, and Ross 1987), and the extent of
agreement among eyewitness experts regarding various findings (Kassin, Ellsworth, and Smith
1989). In general this research has suggested that human memory can be unreliable, and that
witnesses to a crime are usually incapable of providing precise accounts of what they have seen.
A witnesses’ capacity to reconstruct the events of a crime from memory can depend on many
factors: e.g. the age or maturity of a witness, the physical abilities of a witness, the state-of-mind
of a witness during the event she is trying to recall, and the length of time that transpired
between the event and when it is described. Research also suggests that systemic factors (e.g.
how a witness is interviewed by authorities, or how jurors are informed concerning the proper
evaluation of eyewitness testimony) can have a profound effect on the possible truth-finding
value of eyewitness testimony.
Despite the growing body of research exposing the fallibility of the human mind in regard
to memory recall, eyewitness testimony continues to be heavily relied upon in the decision
making process in criminal courts (Huff et al. 1996; Lavrakas and Bickman 1975). Rattner
(1883) found that although 84.1 percent of prosecutors and 62.5 percent of judges ranked
eyewitness misidentification as the most frequent cause of error in a criminal trial, they never43
the-less viewed eyewitness identification as the most critical type of evidence that can be brought
to trial. This paradoxical situation is undoubtedly due to the fact that eyewitness identification
leads to the conviction of many guilty defendants, and criminal justice officials often need to use
this type of evidence if criminals are going to be removed from the street. Eyewitness
identification is ‘direct’ evidence that directly links the suspect to the criminal act, as opposed to
‘circumstantial’ evidence, such as fingerprints, that only prove that the suspect had at some point
touched a piece of evidence. Officials, convinced of a suspect’s guilt, are not likely to throw out
what may be their most persuasive evidence in a case just because they know that, in some cases,
innocent suspects have been harmed by erroneous eyewitness testimony. The burden then falls
on the defense attorney to make a thorough pre-trial investigation of the claims of eyewitnesses,
research the various reasons why a witness may have made a mistaken identification, and to
thoroughly examine an eyewitness at trial. Considering the limited resources and expertise
available to most defense attorneys, it is unlikely that such tenacity will occur.
FACTORS ASSOCIATED WITH EYEWITNESS ERROR
Huff et al. (1996), in their review of research concerning the various factors associated
with eyewitness misidentification, suggest that they can be collapsed into three major categories:
(a) psychological, (b) societal and cultural, and (c) systemic.
Psychological Factors
44
Psychological factors emanate from the disturbed emotional state of victims or
bystanders who are subjected to a criminal act. Trauma or shock can diminish an individual’s
already imperfect powers of perception (Loftus 1979). For example, the amount of stress
involved in an eyewitness situation will affect the recall of details (Christianson 1992; Huff
1996). Studies have established that viewers of violent crime recall fewer details than those who
view a less violent crime (Loftus 1979). Individuals caught up in violent crimes can experience
a “weapon focus” which allows them to describe the weapon used in a crime in great detail, but
inhibits them from adequately describing the individual who was holding the weapon (Yant
1991).
Even in less stressful situations eyewitness identification is fallible. Eyewitnesses can
experience the phenomenon of “unconscious transference” in which the mind recalls a vaguely
familiar face to play the role of a face that cannot be clearly recalled (Buckhold and Greenwald
1980; Loftus 1979). In these cases, eyewitnesses, desiring to remember the face of a criminal
they cannot recall, will conger up a substitute face from their memory of a past encounter.
Societal and Cultural Factors
Flaws in eyewitness identification may result from societal and cultural influences.
Societal and cultural expectations are the beliefs held by large numbers of people in a society
within a cultural context. Allport and Postman (1974) suggest that most people commonly make
perceptual judgments based on stereotypes they possess – that witnesses will more readily
identify a suspect who conforms to their predetermined stereotypical notions. For example, if a
witness believes that people with numerous tattoos on their body are more likely to be criminals,
that witness will most likely identify someone in a line-up who has a large number of tattoos on
his body. Racial stereotypes can also effect misidentification. Brigham and Barkowitz (1978)
45
determined that false identifications are more common when both blacks and whites identified
members of the other race, and when individuals have limited experience with persons of the
other race. Scheck et al. (2000) also found that the largest percentage of misidentification
occurred with cross-racial identification – 35 percent of Caucasians misidentifying African
American defendants.
Systemic Factors
System variables were defined over twenty years ago as “variables that are (or potentially
can be) under the direct control of the criminal justice system” (Wells 1978: 1548). These
variables include procedures that can lead to a wrongful conviction. Eyewitness error leading to
erroneous convictions are sometimes created or attenuated due to systemic forces within the
criminal justice process. Eyewitness evidence, just like physical evidence, can be damaged or
contaminated depending on how evidence is collected. Eyewitness research has demonstrated
that certain methods of conducting line-ups are particularly likely to promote false identifications
of innocent suspects by eyewitnesses (Wells et al. 1998; see also “Mishandling of Eyewitnesses”
below). Systemic errors are perhaps the most important types of errors relating to wrongful
conviction because, if these can be corrected, they will mitigate the psychological, societal and
cultural influences, which may affect the validity of eyewitness testimony. In order to reduce
the incidence of systemic error, criminal justice professionals should be trained to recognize
factors associated with eyewitness error in order that identification can be properly gathered,
assessed, and utilized.
MISHANDLING OF EYEWITNESSES
Today, there is no national set of legal guidelines or rules of procedures for obtaining
eyewitness identification that law enforcement investigators must follow (Wells et al. 1998).
46
Until 1967 the Court made no definitive statements concerning the garnering of eyewitness
testimony. It then handed down three landmark decisions in rapid order that sought to establish
effective constitutional guidelines and safeguards governing the admission of this type of
testimony. In U.S. v. Wade (1967) the Court ruled that suspects are entitled to counsel at postindictment
line-ups. (However, the Court has ruled that there is no right to counsel at preindictment
photo-identification procedures [U.S. v. Ash 1973].) On the same day as the Wade
decision, the Court, in Gilbert v. California (1967), ruled that a suspect had been denied his right
to counsel when, 16 days after appointment of counsel, a line-up was held without counsel
present. However, the Court later reduced the potential impact of Wade and Gilbert when it
decided, in the case of Stovall v. Denno (1967), to uphold a conviction where a witness identified
a murder suspect that was brought to bedside at the hospital for identification; although this was
a classic “show-up” procedure, the Court decided that, in cases of extenuating circumstances it is
an appropriate procedure and due process is not denied. What is most troubling to the cause of
reducing mistaken identifications is that the Wade and Gilbert decisions only apply to ‘postindictment’
procedures. Pre-indictment photo-line-ups or show-ups are the most common
identification method used by police (Stinson, et al 1997); therefore no counsel must be present
in the majority of eyewitness identification procedures.
An eyewitness to a crime might be introduced to various identification techniques used
by criminal justice officials in attempts to exactly identify the individual(s) who committed the
crime, and to further put the witness ‘on the record.” The witness may participate in a “showup”,
photo-spread, and/or line-up procedure. All three techniques offer potential contamination
points for eyewitness’ testimony.
47
‘Show-ups’
In “show-up” procedures, eyewitnesses are shown a single suspect and then asked if that
is the individual they saw committing the crime. (This technique differs from a ‘line-up’
procedure where a witness views several people at the same time). During a show-up session a
witnesses may simply view a single photograph of a suspect at the police station, or may view
the actual suspect in a public setting or some type of custodial setting such as a jail cell, the back
of a police car, or standing handcuffed among a group of police officers.
Show-ups are inherently unreliable (Wall 1965), and are more likely to yield false
identifications than properly constructed line-ups (Dekle, Beale, Elliot, and Huneycutt 1996;
Wagenaar and Veefkind 1992; Yarmey, Yarmey, and Yarmey 1996). Buckhout (1977) found
that one-on-one “show-ups” have a 50 percent potential error rate and are inherently biased. He
noted that witnesses are only given the choice of saying yes or no, as opposed to picking one
person out of a line-up of five or six. He further suggests that some witnesses feel a pressure to
make an identification to avoid disappointing a police officer (see also Loftus 1979; Yant 1991).
The Supreme Court in Stovall v. Denno (1967) warned that “show-ups” can be dangerously
suggestive and prejudicial (see also People v. Barad 1936; People v. Kind 1934; State v.
Landeros 1955). The Stovall decision pronounced that, except in rare circumstances, “showups”
should not be used as a identification procedure because “The practice of showing suspects
singly to persons for the purpose of identification, and not as part of a line-up, has been widely
condemned” (1967: 302); in Stovall, in the witness’s hospital room, the defendant was presented
alone and handcuffed to a police officer. In Simmons v. United States (1968), the majority
wrote:
It must be recognized that improper employment of
photographs by police may sometimes cause witnesses to err in
48
identifying criminals. A witness may have obtained only a
brief glimpse of a criminal, or may have seen him under poor
conditions. The danger will be increased if the police display
to the witness only the picture of a single individual who
generally resembles the person he saw, or if they `show him the
pictures of several persons among which the photograph of a
single individual recurs or is in some way emphasized. The
chance of misidentification is also heightened if the police
indicate to a witness that one of the person’s pictured
committed the crime. Regardless of how the initial
misidentification come about, the witness thereafter is apt to
retain in his memory the image of the photograph rather than of
the person actually seen, reducing the trustworthiness of
subsequent line-up or courtroom identification.
Simmons v. United States 1968: 383
Regardless of the language in this opinion, the Court has not ruled out the use of showup.
The Court has stated that eyewitness identification at trial based on pretrial identification
procedures can only be set aside “if the photographic identification procedure was so
impermissibly suggestive as to give rise to a very substantial likelihood of irreparable
misidentification (Simmons v. United States 1968: 384). This leaves it to a defense attorney to
prove that identification procedures used to identify his client were “impermissibly suggestive”;
many defense attorneys do not have the time nor financial resources to investigate the totality of
identification procedures used during the processing of their client. Defense attorneys are
seldom present at their clients’ line-ups, and when they are, they usually do not understand the
factors that affect suggestiveness to witnesses (Stinson et al. 1996); identification procedures,
including line-up composition, instructions, and presentation are rarely documented –
aggravating the defense attorney’s absence at identification procedures (Scheck et al. 2000;
Stinson et al. 1997 Wells et al. 1998). Also, if unethical police officials or prosecutors are
blatantly being “impermissibly suggestive”, they will no doubt try to cover-up their unethical
actions. Even though the inherent dangers of misidentification during show-ups are well
49
documented, they are still used in police practice to secure eyewitness identifications leading to
arrests and convictions.
‘Photo-spreads’ and ‘Line-ups’
“Photo-spreads” are viewing sessions of photographs, usually conducted at a police
station, where witnesses to a crime are asked to look through numerous “mug-shots (photographs
of criminal suspects taken by police when a suspect is booked following an arrest), hoping that
they will find and identify the perpetrator(s). Most identifications of criminal suspects are from
photographs rather than live line-ups (Wells et al. 1998). Instead of staging a line-up after an
arrest, police use photo-spread sessions 60 percent to 70 percent of the time (Scheck et al. 2000).
“Line-ups” are procedures where police ask a suspect to submit to being viewed, in a live
setting, by the witness to a crime. Many police departments have no standardized line-up
instructions, and identification procedures are seldom recorded (Wogalter et al. 1993). Typically
the suspect will stand behind a one-way glass with five to nine other individuals. Proper line-ups
should include only individuals who resemble the personal characteristics of the suspect as
described by the witnesses (Lindsay, Wallbridge and Drennan 1987; Lindsay and Wells 1980;
Wells et al. 1998). Improper line-up or photo spread procedures increase the likelihood that
there will be a mistaken identification. (Huff et al. 1996; Wells et al. 1998)
Line-up and photo-spread procedures are fraught with pitfalls that can move individuals
who are wrongly suspected to the category of wrongly convicted (Huff et al. 1996; Sheck et al.
2000, Wells et al. 1998). For example, research demonstrates that eyewitnesses to crimes might
misidentify a suspect due to a psychological phenomenon called “relative judgment process”
(Wells et al. 1998). “Relative judgment” suggests that eyewitnesses, when viewing a line-up or
photo-spread, will tend to identify the person who most looks like the culprit, if the real culprit is
50
not included among the individuals viewed (Wells 1984). Experiments by Wells (1993)
demonstrate that most people who identify an actual culprit in a culprit-present line-up, will
simply identify another suspect if the actual culprit is not present. In sum, the relative judgment
process becomes most problematic when the actual culprit is not included among the members of
a line-up or photo-spread, because the eyewitness identifies an innocent suspect as the guilty
party
Experience has shown that line-up and photo-spread procedures can be biased and
suggest a specific defendant. For example, at a parole hearing at Angola Prison, an inmate who
had served 30 years for rape presented to the parole board a photograph of the line-up in which
he was identified as the rapist. There were six men in the line-up – but he was the only one who
was handcuffed (Stack and Garbus 1998). Similarly hastily arranged and biased line-ups can be
created where only the suspect matches the characteristics identified by the eyewitness, while the
other members of the line-up bear no resemblance to those characteristics. When line-ups are
conducted in this manner, the prime suspect becomes the only logical choice for the witness.
The Supreme Court has long recognized the bias inherent in these types of pretrial
confrontations. In U.S. v. Wade Justice Brennan noted:
What facts have been disclosed in specific cases about the
conduct of pretrial confrontations for identification illustrate
both the potential for substantial prejudice to the accuse at
that stage and the need for revelation at trial. . . . . In a case
the defendant had been picked out of a line-up of six men, of
which he was the only Oriental. In other cases, a blackhaired
suspect was placed among a group of light-haired
persons, tall suspects have been made to stand with short
suspects, and, in the case were the perpetrator of the crime
was known to be a youth, a suspect under twenty was placed
in a line-up with five other persons, all of whom were forty
or over (1967:1160).
51
The danger inherent in flawed show-up, photo-spread, and line-up procedures are clearly
demonstrated when one considers the case of Delaware v. Pagano (1979). After a series of
armed robberies in Delaware, police constructed a composite drawing of the perpetrator and
distributed it to local newspapers and television stations. An anonymous caller told police that
the man in the drawing looked like a local Catholic priest, Father Vincent Pagano. Ignoring all
research on the dangers of using the show-up procedure, police took the witnesses to a club that
Father Pagano frequented. When he left the club the police moved the witnesses closer to him so
that the witnesses could get a good look at him. After the show-up, police conducted two photospread
sessions to determine if the witnesses could identify Pagano as the culprit; on both
occasions the background of Pagano’s photograph differed from that of the other subjects, as did
his hairstyle, clothing, and age. To cap off their investigation, police had Pagano submit to a
line-up in which all but one of the witnesses (who had now seen him at the club and in two
photo-spreads) positively identified him as the armed robber. Father Pagano was indicted on
several counts of armed robbery despite his protests of innocence and the lack of any other
evidence against him. At his trial every witness who was called positively identified him, in
front of a jury, as the culprit accused of the crimes. (This is not surprising since the witnesses
had now seen him on numerous occasions over recent months.) Fortunately for Father Pagano,
the criminal justice machinery was halted when, during his trial, another person confessed to the
crimes. The charges against Father Pagano were dropped and the prosecutor apologized in court
(Huff et al. 1996).
The handling of Pagano’s case, from investigation to trial, offers a classic picture of how
not to handle eyewitness testimony. Show-ups should only be used in extreme circumstances
when, for example, a victim is ill or wounded and may be dying. Photo-spreads should only
52
include photographs of people who all generally match the description given by the eyewitness,
and no one photographic background should differ from that of the others (Simmons v. United
States 1968; Stovall v. Denno 1967). Finally line-ups should be conducted so nonverbal cues
that might weigh the identification toward one particular suspect are minimized or eliminated.
Judge and Jury Evaluations of Eyewitness Testimony
After an eyewitness identifies a suspect in a show-up, photo-spread, or line-up, the
suspect is often brought to trial. This is especially true in the case of the wrongfully accused
individual who will often refuse to accept a plea bargain offer. At trial, various safeguards are
presumed to be in-place to prevent miscarriages of justice that result from eyewitness error
(opportunities to suppress evidence, provide expert testimony, cross-examine hostile witnesses).
These safeguards may fail to provide the intended protection because judges, jurors, and defense
lawyers are often not particularly adept at evaluating eyewitness procedures (Devenport, Penrod,
and Cutler 1998; Stinson, Devenport, Cutler, and Kravitz 1997; Wells et al. 1998). Research has
shown that judges may not be capable of properly ruling on erroneous eyewitness testimony
garnered from biased eyewitness identification procedures. Stinson et al. (1997) determined that
judges are, at times, insensitive to the various nuances of line-up suggestiveness that may bias a
witness’s testimony; e.g. “foil bias” (the necessarily heterogeneity in a line-up where the other
line-up members [foils] match the suspect on features described by the eyewitness but vary on
features not mentioned in the witness’s description), “instruction bias” (where instructions fail to
inform a witness that the actual culprit may not be in the line-up, or imply that the culprit is
present) and “presentation bias’ (the use of simultaneous line-ups vs. sequential line-ups).
Stinson, Devenport, Cutler, and Kravitz (1997) found that some judges do not know that the rate
of false identification increases when witnesses, viewing a line-up or photo-spread, are not told
53
that the culprit may not be present (see also Cutlet, Penrod, and Martens 1987; Malpass and
Devine 1981) and that fewer false identifications occur with sequential rather than with
simultaneous presentations (Cutler and Penrod 1988; Lindsay, Lea, and Fulton, 1991; Lindsay
and Wells 1985).
Jurors, likewise, have been found to be unaware of factors that influence line-up
suggestiveness and therefore may not be qualified to accurately evaluate the reliability of
eyewitness testimony. Cutler et al. (1988) determined that the eight variables that have been
empirically shown to affect identification accuracy, (disguise, weapon focus, amount of
violence, retention interval, exposure to mug-shots, biased line-up instructions, lineup size, and
fairness of the line-up) had “trivial effect” on jurors inferences compared to that of “witnessconfidence”
which most strongly influenced the jurors perceived likelihood that the
identification was correct (see also Wells et al. 1998).
REDUCING EYEWITNESS ERROR IN PRETRIAL AND TRIAL PROCEDURES
Research indicates that certain systemic changes in criminal justice pretrial and trial
procedures could reduce the chance of erroneous eyewitness identification leading to a wrongful
conviction. Following are some pre-trial and trial recommendations:
Pretrial
Improved pretrial procedure, according to Justice Brennan, “cannot help the guilty avoid
conviction, but can only help assure that the right man has been brought to justice” (U.S. v. Wade
1967: 1164). Police line-ups and photo spreads can be modified to reduce the effects of the
relative judgment process. Research suggests that the six following steps could improve the
54
authenticity of eyewitness testimony and lessen the chance of error that may lead to a wrongful
conviction:
First, witnesses should be warned that the real culprit might not be present (Wells et al.
1998). Empirical data shows that innocent suspects are less likely to be identified if they are
warned that the actual culprit might not be present (Malpass and Devine 1981; Parker and
Caranza 1989; Parker, Haverfield, and Baker-Thomas 1986). This research has demonstrated
that witnesses, if they are not told that the actual culprit might not be present, have a tendency to
identify the subject who best resembles the culprit relative to others in the line-up – this
phenomenon is known as the “relative judgment process” (Wells et al. 1998).
Second, “blank line-ups”, that is, line-ups in which the real culprit is not present, should
be used before the culprit-present line-up. A blank line-up serves as a type of control line-up (or
“lure”) to see if the eyewitness is able to resist the temptation of identifying someone if given an
opportunity. Experimental research has demonstrated that blank line-ups can be used to identify
eyewitnesses who are prone to make mistakes (Wells et al. 1998).
Third, “sequential procedures” should be used to reduce the chance of eyewitness error; a
sequential procedure is one where an eyewitness views one suspect at a time, and decides if that
person is the culprit before seeing the remaining members of the line-up. Evidence supporting
the use of sequential procedures to prevent relative judgment is impressive (Cutler and Penrod
1988; Linsay et al. 1991; Sporer 1993).
Fourth, in order to avoid influences that might propel an eyewitness in the direction of
making a false accusation, a “double-blind” procedure should be used where the individual who
conducts the photo-spread or line-up does not know whom is the actual culprit. Even when
eyewitnesses are shown proper photo-spreads, and when line-ups contain only similar looking
55
individuals, police officers and detectives can still improperly influence eyewitnesses. If
detectives accompanying the eyewitness during the identification procedure have personal
feelings concerning who may be the real culprit, they may unknowingly (or knowingly) give
subtle hints to the eyewitness that could lead the eyewitness to make a false identification – what
Justice Brennan has described as “the vice of suggestion” (United States v. Wade 1967: 1161; see
also, Fanselow and Buckhout 1976; Luus and Wells 1994; Wells and Bradfield 1998; Wells and
Seelau 1995). Wells et al. (1998) tell an illustrative story:
In State v. Washington (1977) a detective secured a
photograph of someone he thought was James
Washington, his prime suspect in a robbery. Unknown to
the detective at the time, however, he had been supplied
with the wrong photo and the photo was actually of
someone else who was not a suspect. He placed what he
thought was a photo of his suspect in position three of a
six-person photospread and showed it to the eyewitness.
Somehow, the detective managed to obtain an
identification of number three, the very person he thought
was the suspect. Later, when he learned of the error, he
secured a picture of James Washington and created a new
photospread with Washington in position number two.
The eyewitness then identified Washington. (1998: 628).
Fifth, “confidence statements” should be obtained from eyewitnesses at the time of their
initial identification. Confidence statements can also be assessed at various points during the
criminal justice process: during crime-scene interviews, after photo-spreads or line-ups, in
depositions and pre-trial proceedings, and in court examinations. It has been demonstrated that
the confidence that an eyewitness expresses in his or her identification during testimony at trial is
the most powerful single determinant as to whether or not a judge or jury will believe that an
accurate identification was made (Cutler et. al. 1988). Research has demonstrated that there can
be dramatic changes in the confidence eyewitnesses place in their observations as time passes –
56
confidence that may be based on post-identification events which have nothing to do with
memory (Wells et al. 1998). For example, research has demonstrated that witnesses’ confidence
will be inflated after simply being told that the person they identified was a suspect in the case
(Wells and Brasfield 1998). Confidence inflation can also occur after police or prosecutors
suggest to the witness that his or her identification matched their assumptions as to who the
culprit is, or after a newspaper article suggests that the crime has been solved; these types of
events can bolster the confidence of the witness even though the assumptions of police officers,
prosecutors, or news-reporters might be wrong. “Confidence malleability” – the tendency for an
eyewitness to become more (or less) confident in his or her identification as a function of events
that occur after an identification – is particularly important because “actors in the legal system
can contaminate the confidence of an eyewitness in ways that can make an eyewitness’s in-court
expression of confidence a meaningless indicator of the eyewitness’s memory” (Well et al. 1998:
624). Confidence statements, therefore, offer some utility as an index of identification accuracy,
and may give a defense attorney notice that extraneous events following the initial identification
may have bolstered a witness’s confidence. If the confidence of eyewitnesses are noticeably
inflated at trial, then the defense attorney should consider the possibility that heightened
confidence came from sources other than the acuteness of the witnesses’ own memories.
Finally, all line-up and photo-spread sessions should be videotaped (Wells et al 1998).
This would allow for an independent record of the entire process (e.g. instructions given to the
eyewitness, a record of how the actual line-up or photo-spread looked, a record of the possible
verbal or nonverbal suggestions that might have flowed from the line-up agent to the witness,
and a record of the witnesses reactions to the line-up) that could then be used at trial by either
party. Ultimately, this procedure would allow representatives of wrongfully accused or
57
convicted individuals a glimpse into the process that may have led to their miscarriage of justice.
Also, it might act as a deterrent to criminal justice actors who would purposely participate in
biased identification procedures.
Trial
Once a case gets to trial, judges and juries carry the responsibility of properly assessing
how eyewitness testimony was obtained, and its value in the justice-seeking process. In Neil v.
Biggers (1972) and Manson v. Braithwaite (1977) the Court stressed five criteria that should be
used by judges and jurors to evaluate eyewitness identification evidence: (a) the degree of the
witness’s attention, (b) the opportunity the witness had to view the offender at the time of the
crime, (c) the accuracy of the witness’s prior description of the offender, (d) the level of certainty
exhibited by the witness at the identification procedure, and (e) the length of time between
identification procedure and the crime. Eyewitness researchers have criticized these criteria on
several grounds; for example, biased line-up procedures can cause witnesses to become
extremely confident of their identification – even though it was erroneous – and accuracy of
description is not a good indicator of accuracy of identification (Wells and Murray 1983).
Regardless of criticism, these criteria remain the primary guidelines used in today’s court trials
(Wells et al 1998).
Ultimately, it is the responsibility of a judge and/or jury to accurately assess the value of
an eyewitness’ testimony. At a minimum, all parties judging a case where eyewitness testimony
is presented should have an opportunity to hear testimony from eyewitness experts concerning
the strengths and weaknesses of eyewitness evidence. These experts could provide information
concerning, for example, how the amount of stress involved in a situation could significantly
affect the perception and memory of an eyewitness (Huff 1996). Today, many states prohibit
58
such expert testimony (Wells et al. 1998) because they believe it will unduly influence a jury,
and in the states where such testimony is allowable, skeptical judges often do not permit it.
Often, this leaves some judges and jurors with little information with which to make educated
judgments as to the value of an eyewitnesses’ testimony.
CONCLUSION
Innocent individuals who are wrongly accused by eyewitnesses of committing crimes
must confront the simple fact that extensive scientific literature has determined that eyewitness
identification is prone to fallibility, and that the criminal justice process can exacerbate the
consequences of imprecise memory recall. Eyewitness testimony is usually made in good faith.
However, we often know nothing about the conditions under which such testimonies are made.
Potentially suggestive and prejudicial show-up procedures continue to be utilized by law
enforcement officials. Pre-indictment photo-spread sessions – the most commonly used
identification procedure – can be conducted in a biased manner and are often performed without
benefit of counsel. Likewise, live line-ups can be conducted in a biased manner that will
increase the likelihood of a wrongful identification. Further exacerbating the situation is that
defense attorneys may not be knowledgeable enough concerning witness identification
procedures to recognize biased procedures that may damage their case. Research has also
demonstrated that judges also may lack the knowledge to recognize biased identification
procedures and will rule against valid suppression requests. Finally, jurors are often
understandably ignorant of proper identification procedures, yet many states prohibit expert
testimony concerning these procedures, and in the states where such testimony is allowed,
59
skeptical judges often do not permit it because they feel the probative value of such testimony is
outweighed by it potential prejudicial affect on the jury.
The criminal justice system cannot operate without eyewitness testimony. Eyewitness
testimony, therefore, cannot and should not be excluded from the fact-finding process used in the
criminal justice system. All parties that are involved in a search for the truth benefit when due
process is provided, and when accurate unbiased eyewitness accounts are properly obtained from
witnesses. The challenge, therefore, is to refine systemic techniques used to obtain and verify
eyewitness testimony so that more guilty defendants, and less innocent defendants, are
convicted. Police officials, prosecutors, and judges should be better trained and educated
concerning proper methods of obtaining eyewitness testimony.
Most trials are not decided solely on eyewitness testimony – but the ones that are create a
greater chance that there will be a miscarriage of justice. The motives of an eyewitness in these
cases should always be the subject of great scrutiny, and it is up to defense lawyers to educate
themselves so that they can properly cross-examine an eyewitness. Questions should be asked
regarding how a photo-spread or line-up session was conducted. Were procedures used that
biased the witness’s identification? Were any suggestions made to the witness before viewing a
line-up? How much trauma did an eyewitness experience during their ordeal? Has there been a
change in the witness’s confidence in their identification of the suspect? The more valid
questions that are asked, the less likely it is that erroneous eyewitness testimony will contribute
to a wrongful conviction.
In United State v. Wade Justice Brennan suggested that better procedures might reduce
the need for a heightened scrutiny of pretrial identification practices by the Court:
Legislative or other regulations, such as those of local
police departments, which eliminate the risks of abuse and
60
unintentional suggestion at line-up proceedings and
impediments to meaningful confrontation at trial may also
remove the basis for regarding the stage as ‘critical.’ But
neither Congress nor the federal authorities have seen fit to
provide a solution (1967: 1164).
61
CHAPTER FOUR
Just when a scientific principle or discovery crosses the line between
the experimental and demonstrable stage is difficult to define.
Somewhere in this twilight zone the evidential force of the
principle must be recognized. . .
Frye v. United States (1923)
It would be unreasonable to conclude that the subject of scientific
testimony must be “known” to a certainty; arguably, there are no
certainties in science.
Daubert v. Merrell Dow Pharmaceuticals, Inc (1993)
WRONGFUL CONVICTION
AND FAULTY FORENSIC IDENTIFICATION SCIENCE
Faulty forensic identification science, in the context of wrongful conviction, refers to the
presentation to a judge and/or jury, by a forensic ‘expert,’ of mistaken and damning scientific
evidence against an innocent accused. Such evidence can include erroneous blood, semen,
fingerprint, handwriting, ballistics, or fiber analysis, and can be presented both unintentionally
and intentionally by forensic experts. Saks notes, “The testimony some [experts] give is replete
with invisible assumptions and guesswork” (1991: 362). Faulty forensic identification is
particularly harmful to an innocent suspect because judges and jurors, in the same way that they
give great weight to eyewitness testimony, also give great weight to the testimony of laboratory
scientists who swear to the accuracy of scientific evidence presented by the prosecution against
an innocent suspect. Often the testimony of a laboratory scientist can mean the difference
between conviction and acquittal to a wrongly accused suspect (McCloskey 1989). Numerous
researchers have recognized the negative impact of faulty forensic science on wrongful
conviction (Bedau and Radelet 1987; Block 1963; Borchard 1932; Gardner 1952; Huff et al.
1986; McCloskey 1989; Radelet, Bedau, and Putnam 1992; Radin 1964; Scheck et al. 2000;
Yant 1991).
62
McCloskey suggests, “the work of forensic technicians in police crime laboratories is
plagued by uneven training and questionable objectivity” (1989: 58), and that results of
laboratory tests are not always what they appear to be. He states, “ we see instance after instance
where the prosecutor’s crime laboratory experts cross the line from science to advocacy. They
exaggerate the results of their analysis of hairs, fibers, blood, or semen in such a manner that it is
absolutely devastating to the defendant” (1989: 58). McCloskey notes that lawyers often make
no effort to investigate the validity of the expert’s testimony or challenge the findings –
sometimes because they lack the resources to perform needed tests, and sometimes because they
are lazy and incompetent.
Yant (1991) shares McCloskey’s fear of faulty forensic science infiltrating the courtroom.
He suggests that police often blindly rely on the results of reports from crime labs that have no
licensing standards – reports that can contain misleading data resulting from negligence,
incompetence, and outright bias (Conners et al. 1996; Scheck et al. 2000). Yant also criticizes
police for sometimes relying on the nonscientific results of lie-detector/polygraph tests that have
been proven as unreliable. (Lykken [1981], for example, found that innocent suspects fail liedetector
tests about 50 percent of the time.) Finally, Yant suggests that police detectives
sometimes use nonscientific “hypnotic interrogation,” even though such methods have been
proven to be an unreliable method of obtaining the truth (see Grunbaum 1985; Rock v. Arkansas
1987).
Forensic science is different than most other sciences because it attempts to identify an
exact individual and place him or her at the scene of a crime. Saks explains:
Individualization is unique to forensic science. Normal
science is concerned with grouping objects and events
into meaningful classes, discovering systemic
relationships among these classes, and developing and
63
testing theoretical explanations for those shared
attributes and relationships. While normal science
looks only between classes, forensic identification
science purposefully ignores “class characteristics” and
looks within classes. While normal science is
concerned with establishing regularities, forensic
science is concerned with exploiting irregularities
among objects within classes. Its central assumption is
that objects possess enough differences that on
adequate inspection one object cannot be mistaken from
another. (1998: 1074).
What follows are brief descriptions of the case of wrongfully convicted Glen Woodall
who was found guilty of a crime he did not commit – due in large part to faulty forensic
identification.
The Case of Fred Zain and Glen Dale Woodall
State Trooper Fred Zain, the man in charge of serology for the State of West Virginia,
was a forensic superstar. He was capable of finding specks of blood or a spot of semen where
his colleagues were unable to find anything. In sum, he was a prosecutor’s dream – a man with
remarkable vision, possessing phenomenal lab techniques, and the unique ability to detect
genetic markers in crime scene evidence that turned hopeless cases into almost sure winners.
Unfortunately for the criminal justice system, Zain’s work was a massive fraud.
The unraveling of Zain’s exploits began in 1992 when the West Virginia State Police
Internal Affairs Unit investigated the case of a wrongfully convicted prisoner who had recently
been exonerated and released – Glen Dale Woodall. Woodall’s case involved two women who
were abducted and raped in early 1987 at a West Virginia shopping mall. The two victims never
got a good look at their attacker, who wore a ski mask. Working near the mall was Glen Dale
Woodall, the groundskeeper of a cemetery, who had reddish-brown hair and a minor criminal
record. Police searched his home but could find no indicting evidence. Woodall protested his
64
innocence. Detectives took a blood and hair sample from him to compare against the attacker’s
semen and a single hair found in one of the victim’s cars.
The weak case against Woodall was bolstered when, after hypnosis, both witnesses now
said they recognized the suspect both by his appearance and a singular scent. The forensic report
concerning Woodall’s blood and hair sample then arrived. State forensic expert Fred Zain
reported he had determined that Woodall was probably the perpetrator of the crimes. Due to the
weakness of the two victim’s “hypnotic” identification of Woodall, the chief evidence in the case
against him was Zain’s “scientific evidence.” Zain testified in court that, based on his analysis of
blood and semen, only six in ten thousand people could have attacked the women and that
Woodall was a member of that narrow group. Based on this powerful evidence and testimony,
Woodall was found guilty (Scheck et al. 2000: 111).
After five year in prison, Woodall’s relatives contacted the Innocence Project and asked
for help in getting a DNA reanalysis of the evidence against Woodall. Eventually the evidence
was retested and DNA analysis revealed that Woodall was not the source of the semen. Also, the
single hair that Zain had testified to as being one of Woodall’s pubic hairs was determined to
bore no similarity to Woodall’s hair. Woodall was released to home confinement, with an
electronic bracelet on his ankle to keep track of him. Ten months later the state agreed that he
should not be retried and dropped the case against him. Four months after the case was dropped,
the state agreed to pay Woodall one million dollars for his wrongful conviction.
An internal affairs investigation was conducted to establish the extent of Zain’s activities,
and also to determine the state’s liability in the Woodall case. The investigation reported that
Zain’s laboratory could not have performed the tests Zain testified to, and even if he had
conducted the tests, his statistics were totally incorrect. In sum, the investigation revealed Zain
65
had made up a story against Woodall to satisfy the prosecutor and increase his own prestige.
And it wasn’t the first time. The then-secret internal report noted, “Our investigation has
revealed that the trooper who falsely reported the laboratory examination results has done so in
other cases on other occasions and may have testified to those false results. . . . . (Scheck et al.
2000: 109).
Further investigation into Zain’s career revealed that, for least ten years, he had given
evidence in hundreds of serious felony cases and had faked data in almost every case (Scheck et
al 2000). His pattern was always the same. Zain would first claim to have performed tests that
he had not performed. Then he would state that the samples he tested conclusively proved
someone’s guilt, even though he had never tested the evidence. Finally he would make sweeping
conclusions regarding the tests he never performed. Sheck et al. note,
Sham passed for science in criminal cases all over the
United States, at times to shocking effect. Zain was a
prosecution witness in twelve states. He was
responsible for evidence in scores of cases in Texas,
where he had taken a position as chief serologist for the
county medical examiner in San Antonio. On the word
of this dangerous fraud, a jury convicted Jack Davis, a
janitor in an apartment complex, of murdering a tenant
whose body he discovered. Zain miraculously came up
with a DNA test showing blood from Davis on carpet
fibers near the body that was otherwise soaked in the
victim’s blood. As word of Zain’s antics in West
Virginia reached Texas, a team of lawyers led by
Stanley Schneider and Gerry Goldstein exposed Zain’s
DNA test as phony and exonerated Davis. The same
thing happened to Gilbert Alejandro. Zain’s
contribution to his wrongful conviction in 1990
consisted of testimony that DNA tests showed semen
on a rape victim’s clothing “could have only originated
from [Alejandro].” In fact, these tests were
inconclusive. Additional tests showed he was innocent.
Four years later, a Texas judge found that Zain knew
that Alejandro had been cleared by the tests but failed
to report them to anyone (2000: 116).
66
Brief History of Legal Tests for the Admissibility of Forensic Identification Evidence
The first reported court decision affirming the propriety of using “skilled witnesses” was
in the case of Folkes v. Chadd (1782). The courts, as far back as the Civil War, used what has
been recently termed as the “commercial marketplace test” to determine if an “expert” was
qualified to give testimony in court (Saks 1998). Under the marketplace test, a witness was
‘qualified’ as an expert if there was a commercial market for the witness’s knowledge, i.e., if a
person could make a living selling his knowledge, then expertise presumably existed. There are
two problems with this definition as it applies to wrongful conviction. First, the marketplace test
cannot distinguish astrology from astrophysics because the market values both of them. Second,
some fields (e.g. forensic science) do not have a real “marketplace” outside of the courtroom.
Due to these problems, it became necessary for the courts to invent a new test that would better
qualify experts to testify in a courtroom.
In Frye v. United States (1923) the emphasis shifted from considering the marketplace
value of a witness’s knowledge, to the value of that knowledge as expressed by the opinion of
other experts in the field. Judge Van Orsdel in Frye stated:
Just when a scientific principle or discovery crosses the line
between the experimental and demonstrable stage is difficult to
define. Somewhere in this twilight zone the evidential force of
the principle must be recognized, and while the courts will go
along way in admitting expert testimony deduced from wellrecognized
scientific principle or discovery, the thing from
which the deduction is made must be sufficiently established to
have gained general acceptance in the particular field in which
it belongs (1923: 1014)
67
Frye, in actuality, is also a marketplace test; the professional and intellectual marketplace
was simply added to the commercial marketplace as being a necessary ingredient to qualifying
one as an “expert.” The problem remained that the Frye test still was incapable of distinguishing
astrophysics from astrology. The Frye test did, however, separate the expertise from the expert,
and created a legal recognition that a body of asserted knowledge existed. Inadvertently, the
Frye decision transferred control of what was considered ‘valid’ testimony from the commercial
marketplace, where buyers were at least making independent decisions as to the value of the
purported expertise, to the people who produced the knowledge and offered it (and themselves)
to the courts (Saks 1998).
Scheck et al. (2000) suggest that the Supreme Court took steps in Daubert to block “junk
science” from being offered in courtrooms; federal judges would now serve as gatekeepers for
evidence that purported to be scientific. Daubert offers the promise that more guilty individuals
will be convicted and less innocent individuals will be the subjects of miscarriages of justice.
Saks (1998) notes that the shift to Daubert clearly is a shift toward judges making scientific
evaluations more in the manner that scientists themselves evaluate purported science. He
considers this move to represent an increased emphasis on the truth-finding value of trials, but is
skeptical of judges’ ability to “think like scientists” and know what questions to ask to properly
evaluate data. Saks warns that judicial opinions he has recently examined “seemed to use no rule
at all” (1998: 1078), and suggests that judges may ignore rule-guided tests of admissibility.
The Extent of Forensic Identification Error
One might think the case of Fred Zain so peculiar it distorts the real picture of forensic
identification scientist. Consider the forensic misidentifications in San Francisco where,
according to retired police lieutenant Kenneth Blake, prosecutors “had to dispose of about one68
thousand drug cases because of a bad chemist” (Scheck et al. 2000: 119). Or consider the cases
of forensic ‘expert’ Louise Robbins, who specialized in footprint analysis. Her expert testimony
helped convict more than twenty people (Yant 1991). A national panel of 135 anthropologists
and attorneys concluded that footprint analysis “doesn’t work” because of its impreciseness
(Yant 1991: 130); an agent of the FBI crime lab called footprint analysis “totally ridiculous”, and
John Marshall Law School professor Melvin B. Lewis called it “snake oil” (Yant 1990: 130).
Another case of forensic misidentification involved psychiatrist James “Dr. Death” Grigson who
asserted that he could predict, without talking to a patient, how a patient would behave; in
Barefoot v. Estelle the court ruled that Grigson’s questionable assertion was inadmissible at trial
(Yant 1991). The American Psychiatric Association questioned Grigson’s ethical standing in the
psychiatric community. However, despite the fact that his dubious findings led to reversals in
three Supreme Court decisions, Grigson’s testimony was still instrumental in obtaining death
sentences for several Texas defendants (including that of wrongfully convicted Randall Dale
Adams – see chapter 6). Another example of forensic misconduct is found in the case of a
supervisor of forensic toxicology at the Federal Aviation Administration research laboratory,
who plead guilty in federal court to having falsified drug-test results on personnel involved in
train crashes (New York Times, 1991). Lastly, consider the case of forensic scientist Ralph
Erdmann, medical examiner for forty-eight Texas counties, who claimed to have personally
performed four hundred autopsies a year and called himself the “Quincy of the Panhandle.”
Scheck et al. note:
How did Erdman work so fast? By skipping a few
formalities. One family read his death report and was
surprised to see that the weight of the dead man’s gall
bladder and spleen were reported by Erdmann. Years
before the man died, both organs had been surgically
removed. His body was exhumed and showed that no
69
incisions had been done. In another case, Erdmann
claimed to have examined a woman’s brain. Once
again the corpse was exhumed – and there was no sign
that the head had been touched. He became know as
the king of the drive-by autopsy. When a few honest
cops and prosecutors started an investigation, they dug
up seven bodies and found that none of them had
incisions. Erdmann often would want to know the
police theory of the death before he wrote up his report
on the cause of death, the investigators said, to keep his
story straight (2000: 117).
Erdmann was eventually indicted for lying about autopsies he performed. He pleaded no contest
and was given a ten-year suspended sentence. Later his probation was revoked and he was sent
to prison.
Several respected publications have reported cases of fraudulent forensic identification.
Reports concerned: a Texas pathologist who was sentenced for faking autopsies in murder cases
(ABA Journal March 1993); five New York state troopers plead guilty to repeatedly planting
fingerprints in criminal cases (New York Times February 4, 1997); a New Jersey medical
examiner who was convicted of witness tampering in his trial for faking an autopsy report (New
York Times October 29, 1997); a footprint expert who faked results in courtrooms over a 10-year
period (ABA Journal February 1996). Castelle notes, “because it is safe to assume that most of
the forensic science fraud that occurs goes undetected, the amount of fraud that has been
revealed bears disturbing implications for any estimate of the amount of fraud that passes
without notice” (1999: 14).
Fraudulent forensic identification is endemic. McCloskey notes, “the work of forensic
technicians in police crime laboratories is plagued by uneven training and questionable
objectivity” (1989: 58). Scheck et al., in their study of 62 wrongful convictions, note that
‘rigged’ lab tests often contribute to erroneous convictions (2000: xv). They found that
70
‘fraudulent science’ (intentional fabrication of evidence by a forensic specialist) contributed to
33 percent of the incidences of wrongful conviction in the cases they analyzed, and misleading
“microscopic hair comparison” was present in 29 percent of the cases. A study found that 51
percent of police crime laboratories misidentified paint samples, 28 percent misidentified
firearms, and 71 percent erred in conducting blood tests (New York Times Magazine December
11, 1983). Castelle suggests, “Although no one can say for certain [the magnitude] of fraud in
the criminal justice system, there can be little doubt that bias, exaggeration, and error, if not
outright fraud, continues to appear in the work of police labs throughout the country, including
the FBI crime lab” (1999:57).
Inadvertent forensic identification error, exaggeration, and biased results are equally
pervasive in the criminal justice system. Forensic ‘error’ can take the form of inadvertent
mislabeling or switching of samples, mistakenly recording data, mistranscribing results,
misreading or misinterpreting results, and loss of evidence. Inadvertent and inaccurate
“serology inclusion” occurred in 48 percent of the cases 62 wrongful conviction cases studied by
Scheck et al. (2000). Yant notes that a large number of innocent people are wrongfully
“convicted or sentenced to death largely due to the testimony of inexpert ‘expert witnesses’ for
hire, faulty forensic tests, and inaccurate polygraph exams” (1991: 12). Benjamin Grunbaum, a
biochemist at UC Berkley, after studying samples of crime-lab reports, concluded that a
surprising number exhibited “negligence, incompetence, and outright bias” (Yant 1991: 66).
Castelle (1999) notes that even the Zain investigation, which entrusted only the most respected
and independent labs with the reanalysis of Zain’s cases, was haunted with inadvertent errors.
Castelle suggests that it is impossible to know the number of inadvertent forensic errors that are
never discovered, however, he notes, “it is safe to assume that inadvertent error permeates the
71
entire system and can have profoundly tragic results” (1999: 14). It is likely that many of the
“tragic results” have been the wrongful conviction of innocent people.
Reducing Forensic Identification Error
In order to reduce forensic identification error leading to wrongful conviction it is
imperative to understand what type of forensic errors can contaminate a criminal case, and how
these errors are able to effect a miscarriage of justice. For many years forensic identification
error leading to wrongful conviction could only be identified after, for example, the true offender
admitted to a crime, a supposed murder victim turned up alive, or a lying witness confessed.
Now, due to modern DNA technology, researchers have been offered yet another way to better
determine the factors that contribute forensic identification error and wrongful conviction. What
follows is a brief explanation of DNA technology, and how it can be utilized to prevent
miscarriages of justice.
DNA Technology
While the difficulties of studying wrongful conviction have been great, there is now a
new opportunity – a “window-of-opportunity” – that allows researchers to peer into the ‘blackbox’
of wrongful conviction. A new technological phenomenon, DNA (deoxyribonucleic acid)
research, has given criminal justice researchers a moment of rare enlightenment. There is now a
never before present opportunity to determine, in some cases, who has definitely been
wrongfully convicted, and correspondingly, an opportunity to examine those wrongful
convictions cases in order to better determine what factors contributed to these miscarriages of
justice. Post-conviction DNA exonerations therefore provide an extraordinary opportunity to
72
reexamine, with greater insight then ever before, the weaknesses of our criminal justice system
and how they influence the all-important question of factual innocence.
Evidence resulting from the new DNA technology has been widely accepted by the
courts and other agencies within the criminal justice system (Scheck et al. 2000). The new term
‘DNA’ has been linked with a long accepted term ‘fingerprint’, and a new phrase ‘DNA
Fingerprinting’ was coined. DNA fingerprinting (known as restriction fragment length
polymorphism, or RFLP) involves the analysis of blood, skin, or semen evidence that links a
suspect to a crime (or exonerates a suspect) similar to fingerprinting. However, DNA
fingerprinting is much more precise than is old fashion fingerprinting (Scheck et al. 2000).
Until the early 1990’s the RFLP fingerprinting method could not be widely used because
the test required a large amount of DNA for testing. Often, evidence connected to a crime (e.g.
bloody clothing, rape kit, saliva, skin under fingernails) would contain small or trace amounts of
DNA material – rendering RFLP testing useless. Then in 1993 a way to reproduce, or clone,
DNA molecules in the laboratory was invented. A new process polymerase chain reaction, or
PCR, could be conducted using small amounts of DNA evidence which could be reproduced and
made plentiful enough to allow for RFLP testing on all evidence containing DNA materials – no
matter how small the original sample. As a result of PCR technology, old evidence, locked away
in storage and unable to be tested because of the limitations of science, could now be tested. The
results of these tests have often proved that an innocent person was serving time for a crime they
did not commit (Conners et al. 1996; Scheck et al. 2000).
Actually ‘DNA fingerprinting’ is somewhat of a misnomer. DNA research technology is
better understood when it is compared with the long-accepted “blood-type” technology that has
been in use in crime laboratories around the country since the 1930’s. In the past, physical
73
evidence such as blood or semen was analyzed to see if it was from one of the four known bloodgroups:
A, B, AB, and O. If the blood-type found at a crime scene matched that of a suspect, it
could be used in court as evidence against the accused. However, there were many problems
with blood-type technology. It is not a test that makes fine distinctions. Often, there are
thousands of individuals other than the suspect who have the same blood-types and live or work
near a crime scene. Another problem with blood-type evidence is that blood will quickly
deteriorate if not stored properly. Therefore, if blood evidence is not obtained within a short
time after a crime has been committed, it may be of little help to investigators. Also, blood
samples, even if obtained in a timely manner, must be protected from heat, light, and humidity or
bacteria will consume the blood molecules and render the sample useless for blood-typing.
Because DNA molecules are much smaller than the blood antigens that define a person as
type A, B, AB, or O – bacteria will not consume them until the larger blood particles are
consumed. Also, unlike blood-typing, each person’s DNA is unique. Today, DNA can be
extracted from crime-scene evidence and matched to a specific individual. Unlike blood-typing
evidence that must be tested promptly and stored properly, DNA evidence can be found decades
after a crime and still be effectively tested. Scheck, et al. explain the ramifications of this new
technology as well as its limitations;
Now the fabric of false guilt is laid bare …….. Snitches
tell lies. Confessions are coerced or fabricated. Racism
trumps the truth. Lab tests are rigged. Defense lawyers
sleep. Prosecutors lie. DNA testing is to justice what
the telescope is for the stars: not biochemistry, not a
display of the wonders of magnifying optical glass, but
a way to see things as they really are. It is a revelation
machine. And the evidence says that most likely,
thousands of innocent people are in prison, beyond the
reach of the revelation machine, just as there are more
stars beyond the sight of the most powerful telescope.
Most crimes, after all, do not involve biological
74
evidence – blood, semen, hair, skin, or other tissue –
which means there is no genetic material to test (2000,
xv).
This ‘window-of-opportunity’ for taking advantage of DNA research as it applies to
certain types of wrongful conviction will be brief. Since the early 1990’s the new technology
has been widely used. Therefore, in a few years the era of exonerations due to DNA testing will
end. There is now a shrinking population of wrongfully convicted prisoners who can be helped
by DNA testing, and this population will not be replenished because presently many suspects are
now being tested before trial. Of the first 18,000 DNA test results verified by the FBI and other
crime laboratories (over a seven-year period), at least 5,000 suspects were exonerated before
their cases were tried; this means that more than 25% of the prime suspects who were tested
could not be implicated because many, if not most, were completely innocent (Scheck et al.
2000). Due to the fact that the 25% rate remained constant for seven years, and that a 1995
National Institute of Justice’s informal survey of private laboratories (Connors, E. T. Lundregan,
N. Miller, and T. McEwen. 1996) reveals a strikingly similar 26% exclusion rate, there is a
strong implication that the DNA exonerations are tied to some pervasive underlying systemic
problem within the criminal justice process that generates wrongful arrests and convictions. In
past years, without the new DNA technology, many of these individuals might have been tried
and convicted for the crimes they were suspected of committing; only the DNA tests halted the
criminal justice system’s forced march from wrongfully suspected to wrongfully convicted.
Lawyers
Castelle (1999) suggests that defense attorneys need to be more diligent in efforts to
expose and reduce erroneous forensic evidence that may be used against a wrongfully convicted
75
suspect. He notes that defense attorneys too often accept at face value the written forensic
reports from prosecution expert witnesses and fail to schedule a pre-trial hearing to challenge the
admissibility of questionable prosecution science. Defense lawyers should investigate the
underlying data contained in the prosecution’s forensic reports and submit that data for review by
an independent expert in order to obtain independent retesting of any questionable results. Law
students should be better trained to recognize forensic error or fraud, and law and medical
schools, in order to improve the quality of forensic science, should become active sponsors of
first-rate postgraduate forensic science programs. Public defenders offices should have a
minimum of one lawyer who acts full time as a forensic science specialist to help other lawyers
in their cases.
Other Recommendations
The effects of faulty science may also be mitigated if jurors receive ‘innocence training’
before deliberations. This training could inform jurors of the dangers of wrongful conviction
when expert testimony is accepted as infallible.
Forensic fraud could be reduced if crime laboratories, and their budgets, would be
independent from the criminal justice system – and not be beholden to police, prosecutors, or
defense lawyers. All medical labs used to offer forensic evidence should be accredited, subject
to professional standards, and to regulatory oversight. Forensic scientists should be shielded
from non-forensic evidence so as not to contaminate or bias their findings.
Conclusion
76
It has been well recognized that faulty forensic identification directly contributes to
wrongful conviction. Indeed, this has been the very reason numerous studies have been
performed to better identify and hopefully reduce forensic error. What was not originally well
known, but is now well documented, is the degree to which faulty forensic science can indirectly
affect a wrongful conviction by contaminating seemingly independent non-scientific evidence
(Castelle 1999). For example, eyewitnesses who have misidentified a criminal suspect may
become increasingly confident in their identification once they discover that forensic scientific
tests (albeit erroneous) have verified their statement (see State of West Virginia v. Harris 1987;
Castelle 1999: 14). Due to the fact that “witness confidence” has been demonstrated to be a
factor that strongly influences juror’s perceived likelihood that identification testimony is correct
(Cutler 1998; Wells et al. 1998), faulty forensic identification science therefore becomes
increasingly dangerous because it may influence and encourage other participants in a trial
(witnesses, police officers, prosecutors) to more strongly rely on their own erroneous
conclusions.
The damage done by faulty forensic science not only adversely affects the trial of
wrongfully suspected individuals, but it can continue to plague them when their cases are being
appealed. Even when faulty scientific evidence is discovered to have been offered at trial,
prosecutors will typically say that the evidence amounts to a harmless error, and that the
conviction would have occurred even if the faulty evidence had not been presented. Prosecutors
will suggest that other evidence in the trial, e.g. eyewitness testimony, overheard confessions by
a jail house snitch, the lack of an provable alibi, was enough to legitimize the conviction. What
is often unrecognized (and sometime unrecognizable) at the appellate level is the effect the
erroneous expert testimony had on the other witnesses in the trial. Scheck et al. note, “For an
77
innocent person, the two most dangerous words in the language of law are ‘harmless error’
(2000: 172). They suggest that appellate courts use the concept of harmless error to absolve
police officers and prosecutors of misconduct – and expert witnesses of their responsibility in
sending innocent individuals to prison.
Saks (1998) suggests:
The very promise of scientific evidence – that it can
commandingly resolve vital and troubling factual
uncertainties, such as the identity of the perpetrator of a
crime – heightens two competing fears that judicial
gatekeepers must harbor. One is the fear of permitting
rubbish to infect the trial, mislead the jury, and lead to
the wrongful conviction of an innocent person.
Counterposing that is the fear of overlooking evidence
that would forcefully reveal the guilt of a true
perpetrator of a serious crime. Thus, the central
dilemma of trials, namely, the countervailing fears of
convicting the innocent or acquitting the guilty, is
reduced to an evidentiary decision (1998: 1078).
Saks comments appear to take the plight of the wrongfully accused the full circle. From
the pre-Civil War courts, through Frye and Daubert, the fate of the innocent suspect still
ultimately rests on an “evidentiary decision” made by a judge who is usually trying to do the
right thing – that is, see that the guilty are punished and the innocent are exonerated. However,
the experience of history, both ancient and modern, attests to the fact that, in courtrooms, the
guilty are still exonerated and the innocent still convicted. Hopefully, the Daubert test will
reduce the incidence of wrongful conviction, but Saks makes it clear that it will not eliminated it.
Expert testimony will therefore continue to be both a redeemer and a curse to the wrongfully
accused.
78
CHAPTER FIVE
WRONGFUL CONVICTION AND POLICE ERROR OR MISCONDUCT
Far too many cases come from the states to the Supreme Court presenting
dismal pictures of official lawlessness, of illegal searches and seizures, illegal
detentions attended by prolonged interrogation and coerced admissions of guilt,
of denial of counsel, and downright brutality.
William J. Brennan, The Bill of Rights and the States, 1961
It is because the police are typically the first representatives of the criminal justice system
to arrive at the scene of a crime that they figure so prominently in wrongful conviction. Yant
(1991) notes that police error and misconduct are a major cause of wrongful conviction. In his
words, “That [police error] is where much of the problem lies. If innocent people aren’t charged,
they can’t be convicted” (1991: 218). Innocent people who are either “not” suspects, or are only
“possible” suspects to a crime, can quickly become “prime” suspects if police initially mishandle
the case. The mishandling of criminal cases by police officials can be both unintentional (police
error) and intentional (police misconduct). Unintentional police errors known to be associated
with wrongful conviction may include: mistaken arrest, failure to read suspects their Miranda
rights, misreading a search-warrant, or making an unintentional misapplication of the law. Radin
(1964: 17) attributes police error which contributes to wrongful conviction to, “conducting inept
investigations, jumping to conclusions rather than letting the facts lead the way, being to lazy to
check out all the information thoroughly, or ignoring facts which tend to support the story told
by the accused.” Intentional police misconduct involves unethical behavior that may include:
failure to read suspects their Miranda rights, illegal detention, using unduly suggestive
eyewitness identification procedures (e.g. poor show-up, photo-spread, or line-up procedures),
79
intimidation of suspects or witnesses, coerced confessions, fabricating and/or planting evidence,
and suppression of exculpatory evidence.
What follows is the story of wrongfully convicted Steven Linscott who was imprisoned
for a murder he did not commit – due in large part to police misconduct and error.
A Murder and A Dream – The Case of Steven Linscott
Steve Linscott can personally attest to the tragedy that can occur when police officials
arrest and charge someone with murder based only on a “dream” and other inconclusive
circumstantial evidence. One afternoon two police officers knocked on Linscott’s apartment
door and informed him and his wife about the murder of a 24-year-old nursing student in their
apartment building. Needing to make an arrest, police asked them for any information that might
give them a lead. At that time neither Linscott nor his wife could think of anything that might
help the police. As the police left they made a final comment – they asked either of them to call
if they could think of anything, no matter how silly it seemed, that could possibly help them.
Linscott later stated, “Were it not for that parting phrase, their visit would likely have joined the
ranks of the insignificant. But as soon as I heard the words ‘no matter how silly it seems,’ a
dream I’d had the night before jumped into my mind” (Linscott and Frame 1994:19). The
dream, according to Linscott, was ‘unusually intense and most unpleasant’ and involved a man
who was brutally beating another individual.
When Linscott told the police detective about the dream the detective laughed and asked
him to make a written account of the dream. Later the police came to his apartment to pick up
the written account and told him that they would have experts look at it to see if the dream could
help them. The next day police called Linscott and told him they had been in touch with a
psychic, and that the psychic thought the dream could be relevant to solving the crime. They
80
asked him to come to the police station so they could try to get a clearer picture of the killer from
the dream. Linscott scheduled an appointment for the following day. When Linscott arrived at
the police station he was read his Miranda rights. The police officer told him that anyone who
made a statement to police was routinely read his rights. He was questioned for three hours. The
next day a detective again called and asked him to come in again to help with a composite sketch
of the killer. Linscott asked the detective if he was being considered as a suspect. The detective
assured him that he was not a suspect.
Linscott went to the police station the next day. After asking him many questions the
police artist came up with a sketch of the killer that – looked like Linscott. According to
Linscott, “He would later add sideburns and glasses to make the resemblance even closer”
(Linscott and Frame 1994:29). Linscott was then asked to provide blood, saliva, and hair
samples – in order to eliminate him as a suspect. Also, to eliminate him as a suspect, they asked
to look at his car. Linscott agreed. Linscott was then fingerprinted. The detective finally
informed him that he was a suspect – the prime suspect; he stated, “I’d like to think that you are
a person who knows that he needs help. My partner and I and everyone else are convinced that
you killed the young lady, okay? The evidence you gave us tonight will convict you . . . . . You
are going to get the electric chair” (Linscott and Frame 1994:33). Linscott denied any
involvement in the crime.
Based on the “dream” and other circumstantial evidence presented against Linscott, a
jury found him ‘not-guilty’ of rape, but ‘guilty’ of murder. Linscott was sentenced to forty years
in a maximum-security prison. He spent three and a half years in prison before he was released
on bond while awaiting the outcome of his appeal. For the next seven years the prosecutor
pursued a retrial. Finally, in 1992, the state agreed to use new DNA technology to conduct an
81
analysis of the hair and vaginal swap evidence presented against Linscott. As a result of these
tests Linscott was exonerated and the State of Illinois dropped all charges.
This case again shows the multiplicity of factors that can lead to a wrongful conviction.
The police, needing an arrest for a brutal murder, rushed to judgment against Linscott and
arrested him on totally non-conclusive circumstantial evidence. As is typical in cases of
wrongful conviction, once police convinced themselves that Linscott was the murderer, they
stopped following any other leads that might have led them to the real killer – and the real
perpetrator of this brutal crime was left free to continue harming other people. Lies and long
interrogations by the police produced the most flimsy of cases against Linscott. A journalist
speculated years later that police had a vested interested in arresting a white person, given the
social and political tensions associated with integration at that time in the area of the murder
(Linscott and Frame 1994). At trial, the state’s expert witnesses did not fabricate evidence, but
did fail to impress upon the jury the inconclusive nature of the blood and hair evidence presented
against Linscott. The prosecutor withheld exculpatory evidence from the defense attorney and
unethically twisted the words of the state’s forensic expert in his trial summation. Linscott
contributed to his arrest due to his naivety. This is often the case in wrongful convictions.
Suspects, inexperienced and knowing they have done nothing wrong, can become putty in the
hands of experienced criminal justice officials who, convinced of their guilt, will guide them to
arrest and conviction.
Police ‘Error’ and Wrongful Conviction
Unintentional errors by police officers can lead to wrongful arrest. Police, when they
come into contact with individuals at a traffic stop, or while investigating some other type of
misdemeanor, often rely on information from the FBI’s National Crime Information Center
82
(NCIC). The FBI itself admits that 5 percent of the information in NCIC, (a government
databank containing files of millions of fugitives, stolen vehicles, and criminal histories), is
either incorrect or incomplete (Yant 1991). Yant (1991) describes a case involving Roberto
Hernandez who was pulled over at a traffic stop by California police. When police checked the
NCIC file they discovered that Hernandez was wanted in Chicago for attempted burglary.
Hernandez spent eleven days in jail before police discovered the computer information had
mistaken him with another wanted criminal. He received an apology and was set free; two years
later, however, he was again wrongfully arrested for the same Chicago charge.
Police error leading to wrongful conviction can be caused by “overzealousness.” Huff et
al. (1996: 64), concerning their study of wrongful conviction, note:
if we had to isolate a single ‘system dynamic’ that
pervades a large number of these cases, we would
probably describe it as police and prosecutorial
overzealousness: the anxiety to solve the case, the
ease with which one having such anxiety is willing
to believe, on the slightest evidence of the most
negligible nature, that the culprit is at hand.
Police ‘Misconduct’ and Wrongful Conviction
Police misconduct involves unethical behavior that “is the result of a conscious decisionmaking
process to abuse one’s authority while in a position of public trust” (Byers 2000: 1).
Huff et al, note that wrongful convictions often occur when police are “willing to use improper,
unethical, and illegal means to obtain a conviction when one believes that the person at bar is
guilty” (1996: 64). Examples of this type of misconduct are numerous. For example, between
1992 and 1995, five New York state troopers were sentenced to prison for repeatedly planting
fingerprints in criminal cases they were investigating (Castelle 1999: 13). An officer in Oakland,
California, was charged with falsely accusing several men of possessing drugs and weapons
83
between June 13 and July 3, 2000 (Four Officers Face Charges). Regarding the so-called
‘Rampart Corruption Scandal’ in Los Angeles, California, The Los Angeles Times reported that
L.A. police officer Rafael Perez had admitted to personally participating in unjustified shootings,
beatings, witness intimidation, evidence planting, false arrest, and perjury; Perez also named
other L.A. police officers that he says were with him during these incidents (A Second Rampart
Officer Tells of Corruption). Perez was recently sentenced to prison for stealing eight pounds of
cocaine from the LAPD evidence room, and has agreed to cooperate with authorities to expose
corruption in the LA Police Department. He depicted the Rampart CRASH unit of the LA Police
as an “out-of-control group of officers who routinely broke the law to rack up arrests and impress
supervisors” (Four More LAPD Officers Are Suspended). Perez noted that “some Rampart
officers carried stashes of drugs to plant on suspects – usually gang members – who they
believed were guilty of crimes but who did not have drugs on them at the time they were
stopped” (A Second Rampart Officer Tells of Corruption). As a result of Perez’s admissions,
numerous wrongful criminal convictions have been overturned (See Convictions of 9 More
Voided in Scandal and Ten More Rampart Cases Voided). Los Angeles Police Chief Bernard C.
Parks disclosed that 99 people were believed to have been framed and sent to prison because of
disgraced ex-officer-turned-informant Rafael Perez and his former partners (LAPD Chief Call for
Dismissal of All Rampart Cases). The Los Angeles Times reported that more than 3,000
questionable cases would need to be scrutinized due to the Rampart police scandal (Rampart
Probe May Now Affect Over 3,000 Cases). Between January and April 2000, sixty convictions
were overturned by judges in the wake of the scandal (Thirty L.A. Officers Called to Testify for
Grand Jury).
84
Scheck et al (2000) found ‘police misconduct’ to be the third most prevalent factor
contributing to wrongful conviction in their study of 62 exonerations, and determined that police
misconduct was a factor in 50 percent of the DNA exonerations in their study. Other previous
research confirms that police misconduct and error are major contributors to the wrongful
conviction of the innocent (Block 1953; Frank and Frank 1957; Gardner 1953; Huff et al. 1986;
Huff et al. 1996; McCloskey 1989; Radelet et al. 1992; Radin 1964; Rattner 1983; Yant 1991).
Huff and Rattner (1988) suggest that ‘unethical’ police officers, who bring biases to their jobs,
contribute to wrongful convictions by using such practices as: showing a witness a photograph of
the suspect prior to a lineup; coaching witnesses, and coercing confessions. The major types of
police misconduct found to be associated with wrongful conviction are: suppression of
exculpatory evidence, undue suggestiveness in pre-trial procedures, coercion of witnesses,
coerced confessions, and evidence fabrication. A short discussion follows concerning each type
of misconduct.
Suppression Of Exculpatory Evidence
Suppression of exculpatory evidence refers to the intentional withholding of evidence that
would tend to clear a suspect of guilt or blame. Sheck et al. (2000) found this phenomenon in 36
percent of the cases where police misconduct occurred, and determined that the suppression of
exculpatory evidence was the largest category of police misconduct leading to wrongful
conviction.
An example of evidence suppression leading to wrongful conviction is demonstrated in
the case of Thomas H. Broady. He was convicted of first-degree murder during a robbery and
sentenced to 10-25 years in prison. After spending five years in prison Broady was released
when an appeals court learned that police had withheld, during Broady’s trial, the confession of
85
another man who admitted to the robbery and implicated a third man as his partner and the real
killer (Radelet et al. 1992). There is also the case of Robert Kidd who was sentenced to death
after being convicted of first-degree murder. The California State Supreme Court overturned his
conviction when it was discovered that a police witness had introduced a three-page document at
trial that falsely implied that Kidd had a long series of previous arrests. Subsequent investigation
also revealed that police had photographic evidence that was never produced at trial that would
have exonerated Kidd. At retrial, with all of the evidence properly admitted, Kidd was acquitted.
Undue Suggestiveness In Pre-Trial Procedures
Undue suggestiveness in pre-trial procedures usually involves the improper use of showup,
photo-spread, or line-up identification techniques. Hypnosis of witnesses is also a
dangerously suggestive practice that is forbidden in many states (Scheck et al. 2000). Scheck et
al. (2000) determined that undue suggestiveness in pre-trial procedures was the second largest
category of police misconduct leading to wrongful conviction. They found this phenomenon in
33 percent of the cases where police misconduct occurred. Numerous wrongful convictions have
occurred when eyewitnesses, unsure of their ability to identify a suspect, were given improper
suggestions by police authorities that led to a misidentification.
Coercion Of Witnesses
Police coercion of witnesses, in the case of wrongful conviction, involves the
intimidation of witnesses to force them to falsely testify to facts that the police believe will be
instrumental in gaining a conviction. Sheck et al. (2000) determined that police coercion of
witnesses occurred in 9 percent of the cases where police misconduct occurred. Crank and
86
Caldero (2000) suggest that, especially in emotionally charged homicide cases, police officer
outrage and concerns over the victims and their families can lead police to coerce witnesses.
They describe the case of Anthony Porter who was held 16 years on death row for murder.
Porter was scheduled to be executed in February 1998, but his attorney won a stay of execution
due to Porter’s mental condition. Crank and Caldero state, “Subsequently, the main prosecution
witness stated he had been pressured by police to implicate Mr. Porter. Another man since
admitted to the murder, and a corroborating witness has been found” (Crank and Caldero 2000:
45). On February 5, 1999, Porter was released from prison.
Coerced Confession
Police officers coerce confessions for the same reasons that they coerce witness
testimony; they believe that they have “got the right man”, and want to make sure that their
efforts result in a conviction. The Wickersham Commission (1931) published 14 volumes, two
that were about the police. One of the two volumes focused on police misconduct and identified
widespread police abuses to obtain coerced confessions. Huff et al. note,
Among policy makers, police, and prosecutors, use of
[coercion] has been justified, although seldom publicly
acknowledged, when they have been certain that the person
in custody was guilty, and they required a statement from
him or her in order to apprehend accomplices and to obtain
enough admissible evidence so that a jury would bring in a
verdict or guilt against a vicious perpetrator. The
motivation here is to ‘help the jury out’ (1996: 110).
Sheck et al. (2000) determined that police coerced confessions occurred in 9 percent of the cases
of police misconduct they studied. Bedau and Radelet (1987), in their study of wrongful
conviction found misconduct by the police accounted for nearly 25% of errors identified –
usually coerced confessions.
87
The case of Betty Tyson illustrates the danger of police coercion to an innocent suspect.
Tyson was convicted of murdering a well-to-do middle-aged businessman. Although there was
no physical evidence linking Tyson to the crime, the case against her was strong; several
witnesses placed her at the scene of the crime – and she had confessed to committing the crime.
Tyson received a 25-year sentence for the murder. Crank and Caldero note,
A reporter would later find that much of the case was
falsified. The ‘confession’ occurred from a severe beating
she received at the hands of police. One of the witnesses
stated that ‘Detective Mahoney’ had actually put a revolver
to his head and forced him to lie on the witness stand.’ In
another document, one of the witnesses stated that he had
never seen Ms. Tyson at the murder scene, the opposite of
what he stated at her trial.” (2000: 46).
In May of 1999, Betty Tyson, who had become New York’s longest serving female inmate, was
released from prison. Subsequently the State of New York paid her one-million dollars for false
imprisonment.
Evidence Fabrication.
Evidence fabrication also results from the pursuit of the noble-cause, i.e. to-get-the-badguys-
off-the-streets. Thinking that they have arrested a ‘bad’ individual, some police officials
feel that it is necessary to grease the wheels of justice and provide added evidence against the
accused to assure a conviction. Scheck et al. (2000) determined that evidence fabrication
transpired in nine percent of the cases where police misconduct occurred. James McCloskey,
founder and director of Centurion Ministries, Inc., whose efforts have freed numerous innocent
individuals from prison in the past twenty years, lists “perjury by police” as a significant
contributor to the wrongful conviction of the innocent. McCloskey suggests that “what would
88
surprise and even shock most jury members is the extent to which police officers lie on the stand
to reinforce the prosecution and not jeopardize their own standing within their own particular law
enforcement community: (1989:54).
Solutions
Several researchers have concluded that there are steps that can be taken by law
enforcement officers to reduce the incidence of wrongful conviction. Huff et al (1996) suggest
that training of law enforcement personnel should include consideration of wrongful conviction,
including its causes, and methods for prevention. They also suggest law enforcement
performance should be routinely evaluated by state attorney general’s offices, and that law
enforcement personnel who participate in misconduct should be subjected to the most severe
professional, civil, and (were applicable) criminal penalties. In order to reduce the number of
wrongful convictions, Yant suggests that police departments do several things: obtain, and
maintain, a better mix of officers who can look at problems from a variety of perspectives;
improve the accuracy of criminal records; emphasize to police officers that previous arrests
and/or convictions do not mean automatic guilt; make crime labs independent of police
departments (where police detectives cannot influence forensic investigation and where the
presumption of guilt is so predominant). Sheck et al. (2000) and Wells et al. (1998) suggest: all
line-ups, photo-spreads, and other identification processes should be videotaped; police should
be trained about the risks of providing corroborating details that may disguise doubts a witness
might hold; information gained by police officials from jail-house snitches should be subjected
to heightened scrutiny; crime labs should be separated from police influence and act as an
89
independent force in the criminal justice system; and federal involvement should be enhanced in
prosecuting the misconduct of state police officers.
Conclusion
Police error and misconduct are significant contributors to wrongful conviction. Some
police error is unavoidable. Such error can be reduced through better police hiring criteria,
training, and education. Police misconduct is avoidable and should not be tolerated by any
police official. The police become involved in a criminal case at a very critical time – the
beginning. The activities of police officials at this juncture, and how well they do their job, may
have dramatic implications to an innocent individual who becomes a suspect. Once police error
or misconduct contributes to a wrongful arrest, there is an increased likelihood that other
criminal justice officials will add momentum to the mistake. Most of the police misconduct
involved in wrongful conviction stems from attempts to gain status by making arrests or closing
cases, or due to the noble-cause syndrome. In order to better assure a conviction, unethical
police officials are most prone to withhold exculpatory evidence and use biased identification
procedures; to as lesser degree they will fabricate evidence or coerce a witness or the suspect.
Some systemic changes can reduce wrongful conviction, e.g. videotaping identification
procedures and separating crime laboratories from police influence. But it is the police officers
themselves that can do the most to reduce wrongful conviction. All police training should
include instruction concerning wrongful conviction – its causes and how it can be prevented.
Police officers, properly trained, can be educated to understand that preventing wrongful
conviction is also a noble-cause.
90
CHAPTER SIX
WRONGFUL CONVICTION AND PROSECUTORIAL MISCONDUCT
[A district attorney] may prosecute with earnestness and vigor – indeed he
should do so. But, while he may strike hard blows, he is not at liberty to
strike foul ones. It is as much his duty to refrain from improper methods
calculated to produce a wrongful conviction as it is to use legitimate
means to bring about a just one.
Justice George Sutherland, for the majority
Berger v. United States, 1935
The function of the prosecutor under the Federal Constitution is not to tack
as many skins of victims as possible to the wall. His function is to vindicate
the right people as expressed in the laws and give those accused of a crime
a fair trial.
Justice William O. Douglas
Donnelly v. DeChristoforo, 1974
Research has indicated that prosecutorial error and misconduct are major problems
associated with wrongful conviction (Block 1953; Frank and Frank 1957; Gardner 1953; Huff et
al. 1986; Huff et al. 1996; McCloskey 1989; Radelet et al. 1992; Radin 1964; Rattner 1983;
Scheck et al. 2000; Yant 1991). A comprehensive accounting of all cases in the United States
linked to official misconduct and error is not possible because no agency in the country monitors
ethical and legal violations by prosecutors. Appellate opinions do not include those cases that
are not appealed – such as those that are dismissed or plea-bargained. Newspapers only report
the sensational cases. The actual extent of the problem, therefore, is unknown. Humes notes,
There is no reliable or complete source of data on the
total number of individuals released from prosecution or
prison due to official misconduct. Primarily through
press reports and reported appellate decisions, the author
has been able to identify more than one hundred major
felony cases around the country that were undone by
prosecutorial misconduct [between July 1992 and June
1993.] Of course, this represents only a tiny fraction of
the nation’s felony convictions in this period, most of
91
which were untainted and fairly won…… still the
magnitude of the problem is greater than many in the
justice system wish to admit…… These cases represent
only a sampling, for they did not include other instances
of misconduct litigated solely in state courts, where most
appeals die, or those cases in which misconduct may have
occurred, but – because they were resolved by guilty
pleas – were never appealed at all (1999: 587).
In the past several years there has been significant research that indicates that
prosecutorial misconduct leading to wrongful conviction is pervasive (Conners et al. 1996;
Scheck et al. 2000). Yant (1991) suggest there are two major causes of prosecutorial
misconduct: the adversary nature of the American courts that turn searches for truth into
competitive contests between prosecutor and defense counsel, and the elective nature of the
office of prosecutor which compels prosecutors, in order to get reelected, to maintain high ratios
of convictions to indictments.
The mishandling of criminal cases by prosecutors can be both unintentional
(prosecutorial error) and intentional (prosecutorial misconduct). Prosecutorial ‘error’ refers to
those actions where a prosecutor makes a honest mistake that may lead to the wrongful
prosecution and conviction of an innocent suspect. Prosecutorial errors may occur when
prosecutors unknowingly use false or erroneous witness testimony, false or erroneous forensic
evidence, or false confessions. Prosecutorial ‘misconduct’ may include activities in which a
prosecutor, in order to gain a conviction, intentionally withholds exculpatory evidence, fabricates
evidence, coerces witnesses, knowingly uses false testimony, or applies undue plea-bargaining
pressure that may force a suspect to plead guilt to crime he did not commit.
What follow is the story of James Richardson who was wrongfully convicted and
imprisoned for the murder of his seven children – due in a large part to prosecutorial misconduct.
92
Poor, Black, and Victimized – The Case of James Richardson
“Free at last, free at last. Thank God almighty, I’m free at last!” were the words tearfully
shouted by James Richardson on April 26, 1989, as he was released after wrongfully serving 21
years in a Florida prison (Man Revels in His Freedom). A day earlier Richardson had been freed
by a Florida judge who had ruled that Richardson’s trial had been tainted by prosecutorial
misconduct and perjured testimony (Huff et al. 1996). Twenty-one years earlier James
Richardson had stood trial for the poisoning death of his seven children. At trial Richardson was
convicted and sentenced to death on evidence presented that he had killed his children in order to
collect on an insurance policy. Evidence was also presented that Richardson had confessed to
committing the crime to a fellow inmate while in jail awaiting trial.
Prosecutorial misconduct in the case was eventually revealed when crusading attorney
and author Mark Lane was able to obtain a file containing exculpatory evidence that had been
suppressed during this trial. Lane was able to reveal that the Desoto County prosecutor had
framed Richardson for the murder. The suppressed evidence in the file demonstrated that
Richardson had not killed his children to collect on an insurance policy – because there was no
insurance policy. Evidence also showed that Richardson had not confessed to the crime while he
was incarcerated – as a fellow inmate had previously alleged. Suppressed evidence also
indicated that another person, Richardson’s mentally-ill babysitter, had confessed to the crime
several times to friends and acquaintances. Although the prosecutor had in his hands ample
evidence that it was the mentally-ill baby sitter who had actually laced the children’s food with
pesticide on the day they died, he chose to withhold this exculpatory evidence from the defense
attorney and jury. It was later learned that the prosecutors chose to suppress the evidence
because they had already publicly committed themselves to Richardson’s guilt – and because the
93
woman babysitter was a friend of a Desoto County Sheriff. In order to protect the sheriff’s
friend, the prosecutor did not call three key witnesses for fear of what they might say, but instead
offered perjured testimony from other witnesses (Yant 1991).
Prosecutorial ‘Error’ and Wrongful Conviction
Prosecutorial error (honest mistake) can result from erroneous information provided to
the prosecutor’s office by police authorities. For example, prosecutors may unknowingly decide
to prosecute a suspect based on evidence that has been unethically or illegally obtained by
police; e.g. false eyewitness testimony that resulted from undue suggestion, a coerced false
confession, coerced false witness testimony, or fraudulent forensic evidence. Prosecutors may
also not be aware that police officials have suppressed exculpatory evidence that would
exonerate the accused. This type of prosecutorial error demonstrates the ratification of error
phenomenon described by Huff et al. (1986) where, once an error is made at the lower levels of
the criminal justice system, it tends to be reinforced as a case proceeds through the system.
A great deal of prosecutorial error also results from prosecutorial overzealousness. Judge
Jerome Frank wrote, “An overzealous prosecutor is as great a menace to public safety and
tranquility as a deficient, slothful, or corrupt one” (Radin 1964: 45). Huff et al. (1996) suggest
that a prosecutor’s overzealousness may occur because a prosecutor’s office does not have
adequate funds needed to do proper investigations. Lacking resources, and pressed by caseload
pressure and the need to close cases, a prosecutor may resort to unethical tactics in order to gain
a conviction. A favored method of closing cases, especially weak cases against wrongfully
accused individuals, is the use of plea-bargaining.
94
Plea Bargaining
Many researchers have long questioned the so-called ‘voluntariness’ of the negotiated
plea and have raised questions concerning the coercion of innocent defendants (Brunk 1979;
Yant 1991). Innocent individuals who find themselves charged with a crime, thus, can view
plea-bargaining, as a dangerous practice. Lacking sufficient evidence to guarantee a conviction,
a prosecutor will often initially “over-charging” suspects in order to put significant pressure on
them to plead guilty to a crime they did not commit. Plea bargaining pressures are likely to be
increased when a prosecutor’s case against a defendant is weak. Prosecutors, unsure they will be
able to gain a conviction, will often use the power and influence of their office to make a pleabargain
an attractive option – especially for a defendant with few resources to garner in a battle
against the state. In capital cases the prosecutor may offer to the defendant an agreement not to
request the death penalty, or in non-capital cases, the prosecutor may agree to recommend a
reduction in sentence length – even to time already served. In sum, prosecutors are often in the
position to offer a defendant liberty in exchange for a guilty plea; in capital cases they may be
offering a defendant an option of life or death.
Due to the fact that all wrongfully convicted suspects have not committed the crime they
are accused of, the cases against them (if ethically processed) are inherently weak. Pleabargaining
pressure, therefore, will typically be inflicted on the wrongfully accused. Huff and
Rattner (1988) suggest that plea bargaining abuses, especially the piling on of numerous charges
by the prosecutor in order to up the ante of a defendant going to trial, are responsible for many
wrongful convictions. They note that defendants have been known to plead guilty to something
they did not do rather than take the risk of facing a long prison term (Bordenkircher v. Hayes
95
1978; Gregory, Mowen, and Linder 1978). (The topic of coerced false confessions will be
specifically discussed in a subsequent chapter.)
Arenella (1968) asserts that the plea-bargain process adversely affects innocent
defendants because, during the plea-bargaining process, the defendant’s presumption of
innocence, which should be acknowledged in a jury trial, is replaced by a presumption of guilt in
the pre-trial, or plea-bargaining, stage. Arenella states, “When prosecutors respond to a
likelihood of acquittal by magnifying the pressures to plead guilty, they seem to exhibit a
remarkable disregard for the danger of a false conviction” (1968, 32). In the plea-bargaining
stage a prosecutor is not required to present compelling evidence of factual guilt to an
independent factfinder, but instead makes a personal judgment concerning the defendants guilt or
innocence, and becomes not only prosecutor but also both judge and jury.
Prosecutorial Misconduct and Wrongful Conviction
Prosecutorial misconduct involves the purposeful use of illegal or unethical prosecutorial
tactics in order to gain a conviction. These tactics can include: suppression of exculpatory
evidence, knowingly using false testimony, making false and improper statement to a jury in
closing arguments, coercion of witnesses, and fabrication of evidence (Huff et al 1996; Humes
1999; Scheck et al. 2000; Yant 1991). Why do prosecutors engage in misconduct? A report in
the Chicago Tribune suggests that prosecutors do so because they can (Break Rules, Be
Promoted). According to the report, prosecutors who engage in misconduct are rarely punished.
The story notes that not one prosecutor was ever convicted of a crime in the 381 cases involving
prosecutor misconduct that was reviewed by the Tribune. (Many of these cases involved
incidents where exculpatory evidence was concealed, or evidence was presented that the
prosecutor knew to be false). Not only were these prosecutors not convicted, many of them later
96
saw their careers advance and became judges or district attorneys; one of the prosecutors who
had engaged in misconduct became a congressman.
Compounding the problem of lack of accountability is the increased power imputed to
prosecutors during the current “get-tough” era of crime control. Gershman (1999) who annually
publishes Prosecutorial Misconduct, the seminal legal work on its titled subject states,
[It] becomes inescapably clear that the prosecutor,
for good or ill, is the most powerful figure in the
criminal justice system. To be sure, the judge
exercises considerable power, but only after the
prosecutor has made the critical decision about
whom to charge, whom to punish, and how
severely. And this power to charge, plea bargain,
grant immunity, and coerce evidence is largely
uncontrolled. Second, acts of misconduct by
prosecutors are recurrent, pervasive, and very
serious. Case reports do not adequately describe the
extent of such misconduct because so much of the
prosecutor’s work is conducted secretly and without
supervision (Humes 1999: 529).
Prosecutorial misconduct is handled differently, and more discretely, than any other
transgression in the justice system (except for judicial misconduct which is handled the most
discretely of all). When cases are reversed because of prosecutorial misconduct, the names of
the individual prosecutors are rarely noted in the decision. It is rare when the public finds out
about prosecutorial misconduct and even rarer when it learns the name of the perpetrator (Scheck
et al 2000). What can the hapless victim of prosecutorial misconduct do? Usually, they can do
nothing but complain – and those complaints almost always, ultimately, fall on the deaf ears of
those who could take official action. Wrongly charged defendants cannot sue the prosecutor
because prosecutors, in almost every circumstance, have absolute immunity. In sum, there is no
reasonable check in place to balance the power of the prosecutor with the rights the wrongfully
97
accused. Without proper checks and balances, some prosecutors are able to take advantage of
their situation and circumvent the law in order to gain a conviction.
Crank and Caldero (2000) suggest that prosecutorial misconduct often takes place
because prosecutors are in pursuit of the “noble cause”; often a prosecutor feels that the end
justifies the means if a ‘bad guy’ is removed from the streets. The “means” taken by a
prosecutor is most often the withholding of exculpatory evidence, coercing witnesses, or using
false evidence. According to Crank and Caldero, “Prosecutors themselves are often committed
to the noble cause, and in their zeal to convict, step across the line in order to prove guilt in
courtroom proceedings. Unfortunately they all too many times strike ‘foul blows’ as well as
‘hard ones” (2000, 133).
Huff et al. (1996) suggest, however, that the reasons for prosecutorial misconduct may
not always be noble. Instead, they suggest that misconduct can occur because of bigotry, racism,
or even greed. Misconduct can occur because prosecutors simply want to win – they wish to add
points to their scorecard. They wish to be viewed as a successful and tough prosecutor. In sum,
they often wish to enhance their reputation in order to receive a promotion, more money, more
prestige – and possibly a judgeship. Edwin Borchard, over 70 years ago stated, “It is common
knowledge that the prosecuting technique in the United States is to regard a conviction as a
personal victory calculated to enhance the prestige of the prosecutor; it is the environment in
which they live, with an undiscriminating public clamor for them to stamp out crime and make
short shift of suspects, which often serves to induce them to pin a crime on the person accused”
(1932, 369).
Yant (1991) determined the primary activity engaged in by unethical prosecutors is the
withholding of exculpatory evidence, which if known to the defense attorney or jury, might
98
exonerate a defendant. Prosecutors may also, on occasions, allow witnesses to give knowingly
perjured testimony, or use people of questionable character to testify as to a “jail-house
confession.” Scheck et al. (2000) agree. They found that the major types of prosecutorial
misconduct found to be associated with wrongful conviction are: suppression of exculpatory
evidence, knowingly using false testimony, making false and improper statement to a jury in
closing arguments, coercion of witnesses, and fabrication of evidence. A short discussion
follows concerning each type of misconduct.
Suppression Of Exculpatory Evidence
Suppression of exculpatory evidence by a prosecutor refers to the withholding of
evidence from a defense attorney that could either help the defendant establish innocence or
hinder the prosecutor in obtaining a conviction. McCloskey (1989) lists “prosecutorial
misconduct” as one of the seven major causes of wrongful conviction, and notes, “a common
trait of wrongful convictions is the prosecutor’s habit of suppressing or withholding evidence
which he is obligated to provide to the defendant in the interests of justice and fairness.
Numerous cases of wrongful conviction have occurred because prosecutors failed to provide the
defense with known exculpatory evidence (See Adams v. State 1979; Brady v. Maryland 1963;
Miller v. Pate 1967; Oklahoma v. Miller 1987). Scheck et al. (2000) found that exculpatory
evidence was suppressed by prosecutors in 43 percent of the cases of wrongful conviction in
their study.
The case of Todd Neeley demonstrates how the suppression of exculpatory evidence can
lead to a wrongful conviction. Prosecutorial misconduct in this case was revealed when the press
began receiving calls from outraged neighbors who said that they had told police about another
99
suspect – a young boy who lived in the neighborhood, wore braces, had a history of voyeurism
and exhibitionism, and had even bragged about committing the crime. Public outcry called for
the release of Neeley. When the court learned that the prosecutor had never investigated reports
of another suspect, nor had they informed Neely’s attorney about the reports, Neely, after
spending 92 days in jail, was released pending appeal. Four years later, on June 27, 1990, the
appeals court threw out Neely’s “seriously flawed” conviction. The appeals court ordered a new
trial after determining that prosecutors had withheld critical evidence leading to another suspect
in the attack. A judge wrote, “Had this newly discovered evidence been presented to the court in
the original trial, such evidence would have conclusively prevented the entry of a judgment of
guilt” (Yant 1991: 137).
Knowingly Using False Testimony
Scheck et al. (2000), in their investigation of 62 cases of wrongful conviction, determined
that prosecutors knowingly used false testimony in 22 percent of the cases where prosecutorial
misconduct was present. False testimony refers to the making of a false statement(s) under oath,
when one does not believe that statement to be true. Prosecutors knowingly used false testimony
in a 1997 case involving Lisa Lambert. After serving five years in a Pennsylvania prison for the
murder of a romantic rival, Lambert was declared innocent and set free by U.S. District Court
Judge Stewart Dalzell. Dalzell found that prosecutors in the cases had obstructed justice,
knowingly used perjured testimony, and manufactured evidence of Lambert’s guilt while
suppressing evidence of her innocence – all while permitting the real killer, Lambert’s boyfriend,
to escape with a light sentence as a mere accomplice in exchange for his testimony against
Lambert. Judge Dalzell wrote that the justice system had “lost its soul and almost executed an
100
innocent, abused woman. Its legal edifice now in ashes, we can only hope for a barn raising of
the temple of justice” (Pennsylvania v Lambert 1992).
Making False And Improper Statements To A Jury
Scheck et al. (2000) determined that prosecutors had knowingly made false statements or
improper closing arguments in 16 percent of their 62 cases where prosecutorial misconduct was
present. Making false and improper statements to a jury refers to declarations made by
prosecutors that they know to be untrue. An example of a prosecutor making a false statement in
court is demonstrated in the cases of Patricia Donato. She had a conspiracy and fraud conviction
against her overturned when the U.S. Court of Appeals determined that there had been “severe”
prosecutorial misconduct. Donato was accused by the prosecutor of arranging for the theft of her
own car in order to rid herself of burdensome payments and to collect on an insurance policy.
The actual car thief, in exchange for a lighter sentence, accused Donato of paying him to steal
the car. During closing arguments the prosecutors provided the jury with a false motive for the
crime – that returning the car before the lease expired would cost Donato a great amount of
money, leaving fraud as the only way out. Upon investigation, the court determined that
returning the car early would have been a breakeven proposition for Donato and would had cost
her no out-of-pocket money. The court therefore found that there was no reasonable motive for
her to enter into a conspiracy to have her car stolen; they further determined that the prosecutor
had made-up the motive to mislead the jury (Humes 1999).
Coercion Of Witnesses
101
Scheck et al. (2000) determined that prosecutors knowingly coerced witnesses in 13
percent of their cases where prosecutorial misconduct was present. Coercion of witnesses by
prosecutors refers to the compelling of a witness to falsely testify in a criminal trial – through the
threat of physical force or by some other threats to a witness’s well-being. A review of the case
of Ricardo Guerra illustrates how prosecutorial coercion of witnesses can lead to a wrongful
conviction. Guerra was sentenced to death in 1982 for murdering a Houston police officer.
After fourteen years in prison he was freed by an appeals court due to, what a federal judge
called “outrageous misconduct” by prosecutors and police. The court accused prosecutors of
manipulating evidence to falsely convict Guerra when all physical evidence pointed to another
man that was killed in a shootout with police. The judge accused prosecutors of “threatening and
intimidating witnesses, most of whom were children, and hiding evidence of Guerra’s innocence
while lying in closing arguments about evidence of his guilt” (Humes 1999: 639).
Fabrication Of Evidence
Scheck et al. (2000) found that prosecutors knowingly fabricated evidence in 3 percent of
the cases where prosecutorial misconduct was present. Fabrication of evidence refers to the
intentional manufacturing of false proof of a suspect’s guilt, e.g. planting drugs on a suspect,
presenting false forensic results to a court, transferring fingerprints to a weapon, or producing
altered photographs. The Supreme Court case Miller v. Pate (1967) provides an example of the
type of prosecutorial mischief that can occur when evidence is fabricated. The case involved the
murder of an eight-year-old girl. A prosecutor presented to the jury a pair of red-stained men’s
shorts that were found a short distance from a murder scene. The prosecutor alleged the shorts
belonged to defendant, Lloyd Miller, and that the stains on the shorts consisted of blood – and
102
that the blood was of the same blood-type as that of the victim. The defendant denied that the
shorts were his. The prosecutor refused to let the defense attorney examine the shorts and, based
primarily on this evidence, the accused was convicted of murder. It was later discovered that the
red stain was not blood, but instead was red paint. The prosecutor admitted to having known,
during the trial, that the stains were red paint – a fact that he did not reveal to the defense
attorney or the jury. In a unanimous Supreme Court decision Millers conviction was overturned.
Justice Stewart wrote, “the Fourteenth Amendment cannot tolerate a state conviction obtained by
the knowing use of false evidence” (Miller v. Pate 1966: 7).
Solutions
In order to reduce the effect of prosecutorial error and misconduct on wrongful
convictions Huff et al. suggest (1996): basic training for prosecutors should include
consideration of wrongful conviction, including its causes, and methods for prevention – training
should include detailed case studies of wrongful conviction focusing on witness errors, police
and prosecutorial errors, and official misconduct. They also suggest prosecutor’s performance
should be routinely evaluated by state attorney general’s offices, and that prosecutors who
participate in misconduct should be subjected to the most severe professional, civil, and (were
applicable) criminal penalties. In order to reduce the number of wrongful conviction, Yant
suggests that prosecutor’s offices should be “greatly altered”, and that its investigative
responsibilities be “transformed into a subsidiary of an investigating magistracy similar to those
that exist in almost all industrial nations other than the United States” (1991: 219). Yant notes
that this transformation would allow for a single investigative agency that could independently
103
examine the evidence for the police, prosecution, and defense, with “the declared purpose of the
investigation should be to find out what happened and not to develop a case for either side”
(1991: 219). He suggests that such a system would “be a limitation of the uniquely American
adversarial process that turns trials into time-consuming battles between lawyers over issues that
are greatly irrelevant to the question of guilt or innocence” (1991: 219). Sheck et al. (2000)
suggest: prosecutors should be trained concerning the risks of providing collaborating details that
may disguise a witness’s doubts about their testimony, that the testimony of jail-house snitches
should be subjected to heightened scrutiny, and that specialized blue-ribbon disciplinary
committees should deal with misconduct by prosecutors.
Conclusion
Considerable research regarding wrongful conviction indicates that prosecutorial error
and misconduct is increasing in frequency. Error involving the wrongfully convicted usually
occurs when prosecutors unknowingly receive tainted evidence developed by police agencies.
Error also results from prosecutorial overzealousness which can be manifested in plea-bargaining
arrangements where prosecutors, with weak cases, will exert undue pressure on a suspect to
accept an offer of a reduce charge or punishment. Prosecutorial misconduct usually involves
withholding exculpatory evidence from the court; to a lesser degree, prosecutorial misconduct
involves the coercion of witnesses, knowingly using false testimony, fabricating evidence, and
making false or improper statement to a jury. Prosecutorial error and misconduct that can lead to
wrongful convictions may best be reduced through better training of both new and experienced
prosecutors concerning its causes and methods of prevention. More than ever before, defense
104
attorneys must be aware of the sources of prosecutorial error and misconduct, and approach each
case with due skepticism.
105
CHAPTER SEVEN
WRONGFUL CONVICTION AND INADEQUACY OF COUNSEL
The wrongfully convicted invariably find themselves between
the rock of police and prosecutorial misconduct and the hard
place of an incompetent and irresponsible defense attorney.
James McCloskey (1989)
Director of Centurion Ministries
Defense attorneys carry a lion’s share of the burden of responsibility for wrongful
conviction. When the resources of the state (forensic experts, eyewitnesses for the state, police
officials, prosecutors, and sometimes judges) are gathered against an individual who is wrongly
accused, it is often the defense attorney who stands alone with the accused against these powers.
These forces are formidable not only because of their implied credibility, but also because they
have the financial backing of the government. Law schools attempt to train defense lawyers to
adequately represent their clients in these circumstances, but case histories of wrongful
conviction have demonstrated that attorneys, for a variety of reasons, are sometimes unable to
provide adequate counsel to the wrongly accused. Sheck et al. determined that 27 percent of the
defendants in their study “had subpar or outright incompetent legal help” (2000: 187). Other
research confirms the contribution of inadequate defense lawyers to the incidence of wrongful
conviction (Bedau and Radelet 1987; Borchard 1932; Conners et al. 1996; Huff et al. 1996;
McCloskey 1989; Radelet, Bedau and Putnam 1992; Yant 1991).
The following three vignettes illustrate some of the problems wrongly accused
individuals have experienced due to incompetent attorneys.
Good Samaritan Goes to Death Row – The Case of Larry Hicks
106
Larry Hicks was wrongfully convicted of murder and sentenced to die in the electric
chair. At Hicks’ original trial the only significant evidence against him was the testimony of one
of the women who he had helped move. The other woman changed her story so many times the
prosecutor no longer used her as a witness. During the trial Hicks’ court appointed defense
attorney presented a lackluster defense. He did not challenge any of the woman’s statements,
nor did he object to the prosecution’s failure to present bloodstained clothing claimed to have
belonged to Hicks. The defense lawyer also failed to call an alibi witness who was with Hicks at
the time the murder took place, and he ignored statements by the county coroner that an
argument between the murdered men and the killer did not start until after 6 a.m. the next
morning (a fact also apparently missed by the jury). The original attorney was so incompetent
that he did not even file routine appeals on Hicks’ behalf after the guilty verdict.
Less than two weeks before Hicks’ execution, an appellate attorney was able to have
Hicks’ case reopened. A post-conviction investigation determined that: (a) the knife presented as
evidence against Hicks was not the murder weapon. (b) police found no physical evidence
against Hicks, and lost what other evidence they did find, (c) prosecutors had ignored a report by
the chief homicide detective that the investigation had been totally botched, (d) witnesses had
been threatened not to come forward, (e) one of the victims had not been stabbed in the back as
the prosecutor alleged, and (f) the victims were not killed in the place where Hicks’ accuser had
said the killings took place. Despite the new exculpatory evidence, a politically motivated chief
prosecutor, who hated to admit that a mistake had been made, insisted that the new trial go
forward (Yant 1991). At Hick’s second trial, based on the new evidence, he was found to be notguilty.
107
The Case of Edward Honaker
Then there’s the case of Edward Honaker who was accused of rape and kidnapping in
1984. He produced testimony that he was one hundred miles away when the crime was
committed, but was still found guilty and sentenced to a life term. In his trial the state forensic
expert testified that he had examined the slides of the rapist’s semen and determined that the
sperm belonged to Honaker. After spending several years in prison, Honaker was able to get
DNA analysis of the rape evidence, which proved the rapist’s sperm did not belong to him. It
was also learned, when his case was reviewed, that Honaker had received a vasectomy two years
before the rape was committed – a procedure which made it impossible for him to produce
sperm. Although Honaker’s original defense attorney had been informed of the vasectomy, he
decided, for tactical reasons, not to use that information at the trial. The state’s forensic expert
later swore in an affidavit that, had he known about the vasectomy, he would have testified that
Honaker was not the rapist (Scheck et al. 2000). Based on the DNA evidence, and the fact that
Honaker had a vasectomy, the district attorney urged the governor of Virginia to pardon him.
After spending ten years in prison he was freed by the governor in 1994.
The Case Of Frederick Daye
Attorney incompetence affecting the wrongfully convicted also extends to the appellate
level. Frederick Daye maintained his innocence throughout a rape trial and several years in
prison. While Daye was in prison he attempted to get a DNA analysis of the rape evidence used
to convict him; he was sure the analysis would finally exonerate him. The request for DNA
108
analysis was sent to a post-conviction attorney assigned to Daye. Two years passed, and Daye
wrote the attorney from his prison cell to find out if there were any recent developments in his
case. Scheck et al. note,
It turned out that the post-conviction lawyer didn’t have
the first idea about how to handle the new evidence, so
he did nothing. And by the way, even if he had figured
out the procedure, he had managed to lose the files. At
the same time the court was preparing to destroy the
evidence from the rape case, which would have
eliminated Daye’s ability to prove that he wasn’t part of
the rape (2000: 187).
Fortunately for Daye, another attorney had worked on his appeal before turning it over to his
current attorney. That attorney investigated and was shocked to discover that nothing had been
done with the information she had compiled. She successfully filed a motion to preserve the
evidence and, in effect, seized the case. In a few months Daye was freed – proven innocent by
both the forensic DNA test and the affidavit. When news reporters contacted the appellate
attorney who had sat on the case for two years, they asked him if he was concerned about a
malpractice suit. He shrugged his shoulders and told them that he had just filed for bankruptcy
(Scheck et al. 2000).
Defense Attorney Incompetence or Inadequacy
A defendant’s right to counsel is well established in felony cases (Gideon v. Wainwright
1963), and in cases of minor crimes or misdemeanors where imprisonment is possible
(Argersinger v. Hamlin 1972). In the cases of the wrongfully accused, however, the right-tocounsel
per se is not as important as having counsel that is competent and adequate. McCloskey
109
(1989: 57) succinctly describes the typical interaction between assigned counsel and a
wrongfully accused client,
Communication with the defendant is almost nonexistent.
When it does take place, it is carried on in a
hurried, callous, and dismissive manner. Attempts at
discover are made perfunctorily. Prosecutors are not
pressed for material. Investigation is shallow and narrow,
if conducted at all. Preparation meets minimal standards.
And advocacy at trial is weak. Cross-examination is
superficial and tentative. Physical evidence is left
untested. And forensic experts are not called to rebut
whatever scientific evidence the state introduces through
its criminalists…… There are some outstanding
examples of vigorous and through defense lawyers that
leave no stone unturned, but [they] are a rare and
inspiring sight!
Huff and Rattner (1988) list “inadequacy of counsel’ as a leading factor in cases of
wrongful conviction, and note that in many cases original defense counsel do not adequately
represent the interests of the suspect because of inexperience or inadequate investigative
resources. Yant (1991) asserts that defense attorneys representing the wrongfully accused often
are lazy and ill-prepared for trial. He criticizes attorneys who, without fully investigating claims
of innocence, tend to use plea-bargaining as a standard operating procedure in order to reduce
their workload. He also criticizes defense attorneys for seldom taking the time to properly
challenge forensic evidence offered by the prosecution.
Inadequacy of counsel has been a basis for appealing criminal convictions since 1932
(Powell v. Alabama). In cases involving wrongful convictions, defense attorneys have ineptly or
inadequately represented their innocent clients for numerous reasons, e.g., laziness, sleeping
during the trial, appearing at trial in a drunken state or under the influence of an illegal drug, and
poor preparation (failure to make discover motions, depose witnesses, challenge forensic
110
evidence, etc.). Few criminal cases, however, are overturned on the basis of inadequacy or
incompetence of counsel. Scheck et al. (2000) assert that defense attorneys, in order to be
determined ‘effective’ by the appellate courts, usually must only pass the ‘breath test’, i.e. “if a
mirror fogs up when placed beneath the lawyer’s nostrils, he or she is not ineffective, as a matter
of law” (Scheck et al 2000: 183). Also, attorneys are often reluctant to allege that another
attorney is incompetent. Huff et al note,
The rationale that the original defense counsel, for
whatever reasons, did not adequately represent the
client’s interest in the case makes the appeal
difficult to win. Collegial relationships within the
legal profession, though pitting lawyer against
lawyer as adversaries, generally stop short of
promoting the idea of attacking colleagues for
mishandling cases, just as doctors are not eager to
testify against other doctors, police against police,
and so forth. Lawyers assigned anew, or on
appeal, are not eager to pursue this line to gain
reversals, preferring to characterize as “new”
evidence anything that was previously overlooked
(1996: 77).
What follow is a brief discussion of defense attorney incompetence as it is related to
some of the major factors associated with wrongful conviction.
Defense Attorney Incompetence and the Factors Associated with Wrongful Conviction
Defense attorneys who participate in pre-trial identification procedures are sometimes
unable to recognize damaging and biased actions when they are taking place because the
attorney is unknowledgeable concerning these procedures (Devenport, Penrod, and Cutler
1998; Stinson, Devenport, Cutler, and Kravitz 1997; Wells et al. 1998). Once at trial, due to
poor training or other reasons of incompetency, defense attorneys sometimes do not adequately
111
cross-examine eyewitnesses, nor do they investigate changes that may have occurred in witness
confidence concerning damning statements made by witnesses.
Defense attorneys at times do not adequately challenge the forensic evidence presented against
their clients. West Virginia public defender George Castelle (1999) noted that, if even one
defense lawyer had challenged corrupt forensic expert Fred Zain to present his notes during his
reign of error, the charade he was perpetuating would have come to an end. Saks and Koehler
note, “Lawyers as a group evidence an appalling degree of scientific illiteracy, which ill-equips
them to educate and guide the bench in its decisions on the admissibility of evidence proffered
through expert witnesses” (1991: 364). Castelle (1999) suggests that defense attorneys who
represent wrongfully convicted clients too often accept at face value the written forensic
reports from prosecution expert witnesses, and fail to: (a) schedule pre-trial hearings to
challenge the admissibility of questionable prosecution science, (b) obtain the written notes of
telephone communications between the investigating officers and the lab, (c) submit the
underlying data for review by an independent expert, (d) obtain independent retesting of any
questionable results, (e) consult relevant forensic scientific manuals when funds to obtain
expert testimony is limited, (f) and adequately investigate crime evidence that was suspiciously
“overlooked” by police during their original investigation (e.g. evidence that appeared in
unexpected locations, or was found in amounts ‘just large enough to be testable’).
Issues concerning police and prosecutorial error and misconduct are sometimes not
adequately addressed by defense attorneys who may fail to use pretrial motions to identify
instances of suppression of exculpatory evidence, evidence fabrication, undue suggestiveness
in pre-trial procedures, coercion of witnesses, and coerced confessions.
112
Defense attorneys also are many times too ready to accept a plea-bargain offer made by a
prosecutor who has a weak case against their client.
Defense attorneys representing wrongfully accused individuals often do not adequately
investigate the circumstances and motives behind accusations against their client. Sometimes
the witness against the innocent accused is the individual who actually committed the crime, or
is someone who wishes to receive reward money; occasionally an individual who is vindictive
toward the accused will make a false accusation concerning a crime that never took place (e.g.
rape). Failure to challenge false accusations in court may result from the fact that the defense
attorneys themselves often believe their client is guilty and do not provide them with the
presumption of innocence necessary to receive a fair trial. (A detailed discussion of false
accusations is found in chapter eight).
In a similar manner, defense attorneys may not adequately investigate the
circumstances and motives behind confessions made by their clients – because defense
attorneys sometimes work under the assumption that their clients are guilty. Defense attorneys
often do not vehemently challenge testimony regarding so-called ‘jail-house’ confessions
where another prisoner is given a reduced sentence in exchange for testimony regarding the
suspect’s supposed confession to the crime while awaiting trial. Also, false confessions made
by wrongfully accused individuals may occur due to physical beatings by the police, or due to
extreme plea-bargaining pressures exerted by a prosecutor. (A detailed discussion of false
confessions is found in chapter nine).
Lastly, defense attorneys, like other members of the courtroom work group, are
susceptible to community pressure for a conviction. Defense attorneys, in these circumstances,
113
sometimes evade their sworn duty to defend their client to the best of their ability. (A detailed
discussion of community pressure is found in chapter ten).
Why Wrongful Convictions Result from Inadequate Counsel
Inadequate or incompetent representation by defense attorneys of the wrongfully
accused attorneys may result due to one or more of the following four issues: First, eighty-five
percent of criminal defendants cannot afford an attorney and are either assigned a public
defender or court appointed attorney paid for by the government (McCloskey 1989).
Incompetent counsel, therefore, is not typically chosen by the wrongfully accused, but instead
is assigned to the wrongfully accused by the government. Regarding assigned counsel, Scheck
et al. note,
In the famous case Gideon v. Wainright, the Supreme
Court said every defendant, rich or poor, was entitled to
an attorney. For the most part, the systems of “indigent
defense” that have emerged in the last thirty years
makes no pretense of complying with Gideon’s
mandate of supplying counsel to improve the
defendant’s chances of receiving a fair trial. The poor
person is often assigned a lawyer who lacks knowledge,
skills, or even the spirit to defend the case properly.
(2000: 188)
Due to the fact that eighty-five percent of criminal defendants are indigent, the wrongly
accused typically receive either an attorney from a public defender’s office or have counsel
assigned to them. Public defender offices often have capable defenders but, due to the
overwhelming caseloads and low-level of compensation, the burnout rate is high. McCloskey
(1991: 57) notes, “As competent as full-time salaried public defenders generally are, their
resources (budget and people) are vastly inadequate and are dwarfed by those of their adversaries
114
(the local prosecutor’s office).” The compensation paid to assigned private attorneys who defend
indigent clients is usually also woefully low. Scheck et al. (2000) report that the maximum fee
for a non-death penalty case in Mississippi is $1,000. In some areas of Texas the limit is $800.
In Virginia the most a court-appointed attorney can receive is $305 for defending a client facing
punishment of less than twenty years. The low compensation, of course, typically keeps the
more experienced and competent attorneys from taking the cases of the indigent, resulting in the
cases being distributed to a pool of younger, less experienced attorneys. Another consequence
of low compensation is that, in some cases, it has actually led to profiteering by some defense
attorneys. So-called ‘cop-out lawyers” are able to make a comfortable living by simply pleading
their clients guilty without adequately investigating their cases or even interviewing their clients.
Many wrongfully accused individuals have pleaded guilty to a crime they did not commit
because of pressures exerted on them by defense attorneys and prosecutors. (See ‘false
confessions’ chapter nine.) Silberman (1980) reported that attorney “wholesalers,” who charged
only $50 per case, were able to earn over $50,000 per year by having their clients plead guilty in
five cases per day. Huff et al. (1996) report that one Los Angeles attorney entered as many as 25
guilty pleas per day and received $200 a case; his income was as high as $5,000 per day.
Second, some defense lawyers are poorly trained to protect the innocent individual who is
accused of committing a crime. All basic and advanced training should include consideration of
wrongful conviction, its causes, methods of prevention, and its implications to society. Huff et
al. (1996: 151) suggest, “This training should include the presentation of detailed case studies of
wrongful conviction, focusing on witness errors, police and prosecutorial errors, official
misconduct, and judicial failure to detect errors at trial and on appeal.”
115
Third, defense attorneys who are involved in substance abuse sometimes represent
wrongfully convicted individuals. The State Bar of California admits that thousands of lawyers
in that state alone come to court each day either drunk or high on drugs (Yant 1991). In the early
1990’s the American Bar Association went public with a high-profile campaign to combat
substance abuse among lawyers. Surveys conducted by the bar associations in Oregon and
Washington indicate that 15-18 percent of lawyers are alcoholics. The Texas State Bar estimates
20 percent of its members has a substance abuse problem – twice as high as the general
population (Yant 1991). “We find that in a lot of cases, the person suffering from substance
abuse is not handling business as they should’, said Bill Davis, who handles complaints about
attorneys filed with the ABA” (Yant 1991: 167). Yant (1991) notes that lawyers who turn to
drugs and alcohol usually end up hurting their clients financially, although not before many have
been already been victimized by inadequate counsel
Lastly, another reason incompetent lawyers continue to represent the wrongfully accused
is because such lawyers are usually not disciplined when their conduct is unprofessional and
unethical. Most states have systems of ‘self-regulation’ that are developed, according to the
American Bar Association, “to maintain appropriate standards of professional conduct in order to
protect the public and the administration of justice from lawyers who have demonstrated by their
conduct, that they are unable or unlikely to be able to properly discharge their professional
duties” (ABA Standards for Lawyer Discipline and Disability Proceedings, Standard 1.1, 1979).
ABA standards, however, are often not enforced. Guttman and Bumstead (1986 E-1) note;
Nationally, lawyer discipline is a travesty. Misconduct is
rarely perceived. If perceived, it is not reported. If
reported, it is not investigated. If investigated, violations
are not found. If found, they are excused. If they are not
excused, penalties are light. And if significant penalties are
116
imposed, the lawyer soon returns to practice, in one state or
another.
Even when appellate courts overturn convictions based on misconduct, nothing typically happens
to those who broke the law and their oaths. According to Ramsey (1994: 73), “Self-regulation
has always served primarily as a vehicle to maintain lawyer’s autonomy, economic position, and
established status in society rather than as a vehicle to cleanse the profession of unqualified
members and protect the public.”
Conclusion
Reducing the incidence of wrongful conviction due to inadequate or incompetent defense
counsel might be realized if the compensation for public defenders and assigned private counsel
would be increased so as to attract more experienced, competent, and ethical lawyers. The
salaries of public defenders should be at the same level as that of local prosecutors. At the
appellate level, Scheck et al. (2000) note that the direction of compensation appears to be going
in the opposite direction since Congress has defunded the program that provides money to
appellate attorneys after capital convictions. Public defender caseloads should not exceed
accepted standards that are sanctioned by the National Legal Aid and Defenders Association, and
public defenders should receive increased continuing legal education and more flexible work
plans. Law schools should train attorneys concerning the phenomenon of wrongful conviction so
students can understand its causes and methods of prevention. Local bar associations and
appellate courts should enforce performance standards, and penalties should be applied to public
defenders or private attorneys who fail to meet these standards (Huff et al. 1996). Scheck et al.
117
(2000) suggest that the courts, by failing to find fault with even the worst examples of
incompetence, have ratified such conduct.
118
CHAPTER EIGHT
WRONGFUL CONVICTION AND FALSE ACCUSATIONS
[In some cases] the prosecuting attorney [is] obliged to take the
evidence presented to him, including the uncontrollable perjury of a
vengeful witness, and lay it before the jury without realization of its
worthlessness.
Edwin Borchard Convicting the Innocent 1932
False accusations are distinguished from eyewitness error in that false accusations are
purposeful and perjured untrue statements intended to convict an innocent person. Unless the
false accuser’s lies are exposed, judges and jurors will often give great weight to their testimony,
and may convict an innocent person based solely on this testimony. Scheck et al. (2000)
determined that “false witness testimony” was the sixth most prevalent factor associated with
wrongful conviction. In their study they found that “snitches or informants” gave erroneous, and
often self-serving information that implicated an innocent individual in 21 percent of the cases
they examined. Huff and Rattner (1988) also classify “false accusations” as being a major
problem leading to wrongful convictions. Borchard (1932) in his classic study of convicting the
innocent determined that in 25 percent of the cases he studied there was a “frame-up” involved;
i.e., an individual was wrongfully convicted due to a story that was fabricated by someone else.
Bedau and Radelet (1987) in their study of miscarriages of justice set up the highest incidence of
false accusations; they found false accusations in one-third (117) of the cases they studied –
almost double their findings of innocent eyewitness error (in 57 cases). They note, “This type of
corruption spans the years and the jurisdictions; it is too frequent and too familiar” (1987, 60).
Huff, Rattner, and Sagarin (1996) found “perjury by witness” and “frame up” in approximately
15% of the 205 cases of wrongful conviction they studied.
119
On the Wrong Side of the Thin Blue Line – The Case of Randall Adams
False accusation tragically affected the life of Randall Dale Adams. Adams was
convicted and sentenced to death by a Dallas, Texas, jury for the November 28, 1976 murder of
Dallas police officer Robert Wood. A young man named David Harris linked Adams to the
crime. Adams had met Harris when he ran out of gas and was offered a ride by Harris to a gas
station. Adams ended up spending the day with Harris. Later that evening, after Harris had
dropped off Adams at a motel, Harris murdered Officer Wood when he was pulled over because
his automobile’s headlights were off. It was Harris who was later to frame Randall Dale Adams
for the crime and send Adams to Texas’s death row.
One month after the murder of Officer Woods, detectives learned that Harris was
bragging to his friends about killing a police officer. When police interrogated Harris he
admitted to being at the scene of Wood’s murder, and gave the police the murder weapon.
Harris accused Adams as being the individual who pulled the trigger of the gun that killed
Officer Woods. Adams (who had no criminal record) was then arrested at gunpoint at his place
of employment. He was taken to the police station and subjected to an exhausting interrogation
with no attorney present, no food, no water, and no rest breaks. According to Adams, when the
detectives were unable to get him to confess to the crime, they threw a gun on a table and told
him to “Pick the son of bitch up and look at it!” When Adams would not pick up the gun and put
his fingerprints on it the detective purportedly pulled his service revolver and, according to
Adams, pointed it at Adams’ forehead, cocked it, and said, “I ought to blow your shit away”
(Adams 1991: 28).
The district attorney’s star-witness was David Harris. In exchange for Harris’s
testimony, all charges against him concerning the crime were dropped. The prosecutor and judge
120
withheld from the jury information that Harris had picked up Adams in a stolen car and that,
after the murder of Officer Wood, Harris, by himself, had for the next several weeks, participated
in an extensive crime-spree. Still the prosecutor had a weak case against Adams. In order to
buttress his case, the prosecutor, on the last day of the trial, produced three eyewitnesses who
testified that they were driving by the crime-scene the night Officer Wood was murdered and had
seen Adams sitting behind the wheel of the car that Wood had pulled over. Until this last day of
trial, neither Adams nor his attorney had ever been told that anyone had witnessed the crime.
Stunned by the new testimony of these witnesses Adams’ defense attorney asked for a
continuance and investigated. Three days later the defense attorney attempted to recall all three
witnesses for cross-examination. The prosecutor told the court that the witnesses had left town
and could not be located; it was discovered much later that the witnesses were still in Dallas but
had moved to a different motel. Telephone records later proved that the prosecutor knew where
the witnesses were, but withheld that information from Adams’ attorney (Radelet, et al. 1992).
Adams was convicted due to prosecutorial misconduct and the false accusations of four
individuals – Davis Harris and the three surprise “eyewitnesses.” Harris, who later admitted to
committing the crime alone, stated that he had testified against Adams in exchange for immunity
from prosecution. It was also revealed that the prosecutor’s three ‘surprise’ eyewitnesses had
lied on the witness stand in order to collect the reward money offered to solve Officer Wood’s
murder (Huff et al. 1996). It was also discovered that the prosecutor had failed to turn over to
the defense written affidavits from the three witnesses, and the original police report of the
murder – all of which contained conflicting information that could have exonerated Adams.
In 1989 Randall Adams was pardoned by the Texas governor and released from prison.
121
Pursued for Eight Years – The Case of Tony Cooks
In 1978 Cooks, who had just turned 18, was accused of murdering John Gould. The
forty-two-year-old Gould had been accosted by three black teenage youths one evening while he
and his wife walked down a street near their home. Seeing the crime from her apartment
window, Helen Foster identified Cooks as one of the assailants. Gould’s wife told police that the
assailant was “a light-skinned black.” Police showed her a photo-lineup in which Cooks was the
only light-skinned black, and she told police, “I can’t be positive, but I think that’s him” (Radelet
et al. 1992: 191). Two days after Cooks’ arrest another 14-year-old youth was also arrested after
he confessed to his own involvement in the crime; the young man then falsely accused Cooks as
also being involved in the crime. Based on these eyewitness identifications, Cooks was indicted
for murder.
Cooks’ first trial ended up in a hung jury; his second trial ended in a mistrial; his third
trial ended in a hung jury. Finally, at his fourth trial, Cooks was successfully prosecuted on the
murder charge. The most damaging evidence presented to the jury came from the 14-year-old
youth who pointed to Cooks and identified him as an accomplice to the crime. Also, the victim’s
wife, who had originally stated “I can’t be positive, but I think that’s him,” then positively
identified Cooks as the killer. In his opinion Judge Roosevelt Dorn scoffed at the idea that the
witness Helen Foster could have positively identified anyone at nighttime in the street 177 feet
from her apartment window; he also believed that the victim’s wife might have been misled into
making a false identification due to the biased photo-lineup procedure used by the police. The
prosecutor appealed the judge’s decision and the appellate court reinstated the conviction. The
122
judge, now forced to pronounce sentence, ordered Cooks to prison for sixteen years to life, but
freed him on $5,000 bond pending appeal. On appeal Cooks won his right to a fifth trial.
In 1986, at Cooks’ fifth trial, it was revealed that the case against him was based on the
false identification by a self-confessed participant in the murder, and on mistaken identification
by an eyewitness. Previously uncovered evidence was introduced at the new trial showing that
the 14 year-old eyewitness against Cooks had told his probation officer that his testimony was a
“lie” he made up in order to satisfy a persistent detective who would not take ‘no’ for an answer.
The jury voted to acquit Cooks. His wrongful conviction had been based on: (a) the false
accusation of the 14 year-old youth who testified against him to make things easier for himself,
(b) the acts of an overzealous detective and prosecutor, (c) the suspicious testimony of a woman
who possibly had an axe to grind against Cooks for his previous encounter with his son, and (d)
the mistaken identification of the victim’s wife whose confidence in her original tentative
identification of Cooks was likely bolstered at trial (and made more believable to the jury) when
she discovered that Helen Foster and the 14 year-old youth had identified Cooks as the culprit
who had murdered her husband.
Types of False Accusations
False accusations are always made because the accuser has something to gain (except in
cases where the accuser is mentally-ill). False accusations can originate from numerous sources.
For example, an individual who commits a crime may falsely accuse an innocent person of the
offense in order to cover-up his or her own involvement in the incident (Bedau and Radelet
1987; Radin 1964). Bedau and Radelet (1987) found that the possibility of a false accusation
increases when an accomplice to a crime makes the accusation. False accusations also occur
123
even when no crime was committed (Huff et al. 1996). This type of false accusation can come
from vengeful spouses (due to infidelity or divorce), mistresses (revenge or money), business
partners (to take control of a business), or from acquaintances that believe they can profit, either
emotionally or financially, by seeing an innocent person convicted of a crime (Bedau and
Radelet 1987; Radin 1964). Yant (1991) notes that the two most common reasons uncovered by
research for false accusations of rape are: (a) to explain a problem of pregnancy, venereal
disease, or evidence of promiscuity, and (b) revenge. Some individuals have even admitted to
falsely accusing an individual of a crime “just for fun” (Frank and Frank 1957:193; Rattner
1983). False accusations may also occur so that the accuser can benefit from a civil lawsuit, or
to gain reward money (Huff et al. 1996; see also Adams v. State 1979). False accusations of
child molestation and other sex crimes for financial gain are common, and do a great disservice
to those victims who truly are the subjects of these types of crimes (Yant 1991).
Some types of false accusations are either initiated, or aided and abetted, by criminal
justice officials. For example, police officers have falsely accused innocent individuals in order
to increase their number of “collars,” or to satisfy their own racial or ethnic biases and urges. In
1999, disgraced police officer Rafael Perez of the Los Angeles Police Department admitted to
fabricating evidence in order to convict at least 99 innocent people of crimes they did not commit
(LAPD Chief Calls for Dismissal of All Rampart Cases). Prosecutors have knowingly accused
innocent individuals of committing crimes they did not commit (see Pennsylvania v Lambert
1992). Also, prosecutors have unknowingly accused (and convicted) innocent individuals of
murder when the victims actually had died of natural causes or by accidental death (Radelet et al.
1992). Perhaps the most troubling false accusations that involve criminal justice officials are
those that come from so-called ‘jailhouse snitches.’
124
Jailhouse Snitches
A major source of false accusation comes from criminals who are already locked up in
jails and prisons. Jailhouse snitches are inmates who, in exchange for their false testimony, will
typically be rewarded by the prosecution with a reduction in sentence or outright release. In
order to gain the favor of police and prosecutors, these jailhouse snitches will concoct stories of a
so-called ‘jailhouse confession’ which was purportedly made by an innocent suspect who was
charged with a crime and awaiting trial – and then testify in court to the authenticity of their
claim (McCloskey 1989). False testimonies by jailhouse snitches may also involve the snitch
stating that they saw a defendant fleeing a crime-scene, or that they saw them near a crime-scene
shortly before a crime took place. What makes this type of false accusation so troubling is the
participation of criminal justice officials in the snitch’s scheme; in essence, these officials are
sub-optimized by the jailhouse snitch, and lend creditability and legitimacy to the false
accusations. James McCloskey, advocate for the wrongfully convicted, notes,
If I have seen one, I have seen a hundred “jailhouse
confessions” spring open the prison doors for a witness
who will tell a jury on behalf of the state that a
defendant confessed a crime to him while they shared
the same cell or tier. When the state needs important
help, it goes to its bullpen, the local county jail, and
brings in one of the many ace relievers housed there to
put out the fire. As several of these “jailhouse priests”
have told me, “It’s a matter of survival: either I go away
or he [the defendant] goes away, and I’m not goin’.”
Jailhouse confessions are a total perversion of the truthseeking
process (1989, 55).
Conclusion
False accusations often lead to the conviction of innocent individuals and devastate the
lives of not only the accused, but also that of their families. Many times false accusations only
125
serve to divert suspicion from the truly guilty person. Ultimately, false accusations are tendered
by pernicious perjurers who, for the sole reason of financial or emotional gain, use the criminal
justice process to attain their goals. Sadly, false accusations sometimes occur because it creates
a win-win situation for both the accuser and criminal justice officials; the false-accusers
accomplish their sorted goals, and the criminal justice officials get a conviction. Prosecutors and
police officials may contribute to the incidence of false accusations because they too readily
accept them. Especially in the events of weak cases, where criminal justice authorities need
additional evidence in order to increase their chances of gaining a conviction, they may welcome
anyone who will provide them with testimony that will bolster their case. The use of jailhouse
snitches is possibly the most blatant example of the misuse of false accusations because of the
active role taken by criminal justice authorities in many of these cases. Leslie Abramson, past
president of the California Attorneys for Criminal Justice, notes, “They [police and prosecutors]
shouldn’t be using these people [jailhouse snitches] because they know they’re lying. The real
tragedy is that they only use these guys in cases where the evidence is weak. So they increase
the risks of convicting the innocent” (Yant 1991: 128). Some states have taken steps to reduce
false accusations in capital cases by providing for the death penalty for anyone convicted of
perjury where an innocent defendant was convicted and executed (Bedau and Radelet 1987).
126
CHAPTER NINE
WRONGFUL CONVICTION AND FALSE CONFESSIONS
Confessions have often been extorted to save law
enforcement officials the trouble and effort of obtaining
valid and independent evidence.
Escobedo v. Illinois, United States Supreme Court (1964)
False confessions are especially dangerous to innocent suspects because this type of
evidence is likely to dominate all other case evidence and lead a trier of fact to convict a
defendant. False confessions are universally treated as compelling and damning evidence of
guilt by judges and jurors simply because most suspects who confess are guilty. Scheck et al.
(2000) note that, in cases where false confessions are presented as evidence to juries, 73 percent
of juries will vote for conviction, even when the confession is repudiated by the defendant and
contradicted by the physical evidence. False confessions are one of the least publicized and most
misunderstood dynamics of wrongful convictions. Why, after all, would a perfectly innocent
person plead guilty to a crime he or she did not commit? False confessions occur for numerous
reasons, and can be divided into two categories: coerced and voluntary.
Coerced False Confessions
Coerced police-induced false confessions rank among the most significant of all official
transgressions. In 1931 the Wickersham Commission devoted one volume of its 14-volume
report to police lawlessness and widespread police abuses to obtain coerced confessions. In
1966 the Supreme Court handed down its landmark Miranda v. Arizona decision in which the
Court sought to protect all suspects from acts intended to get them to make confessions against
their will – confessions blatantly in violation of a suspect’s Fifth Amendment protection against
self-incrimination. Yant (1991) notes that false confessions still occur when police forego the
127
Miranda warning and tell suspects they must talk to them. Researchers have documented
numerous cases of police-induced false confessions in the decade of the 1990’s alone (Conners et
al. 1996; Huff et al. 1996; Kassin 1997; McMahon 1995; Ofshe and Leo 1997; Yant 1991).
After innocent suspects are arrested, they will typically protest their innocence to the
police. Due to the fact that police officials tend to assume guilt as a working premise of their
craft, claims of innocence are usually given little creditability (Klockars 1991). Police, assuming
the accused is guilty, may attempt to coerce a confession from a suspect by using one of many
available interrogation techniques. These techniques can utilize, for example: threats, tricks,
manipulation, torture and brutality, exhaustion, starvation, fear, and fatigue (Huff et al. 1996).
Leo and Ofshe (1998) note that police officials often use techniques from influential training
manuals such as Criminal Interrogation and Confessions (Inbau et al. 1986) and Practical
Aspects of Interview and Interrogation (Zulawski and Wicklander 1993), which have been
shown to be coercive and produce false confessions.
If police officials are unable to extract a confession from a suspect, then prosecutors (who
may also be skeptical of a suspect’s claim of innocence) will begin to apply pressure on the
defendant to confess. Prosecutor-induced false confessions are especially prominent in cases
involving the wrongfully accused; due to the fact that the suspect actually did not commit the
crime, prosecutors may lack sufficient evidence to guarantee a conviction. In response to this
dilemma, and to avoid going to trial, prosecutors often apply significant plea-bargaining pressure
in order to persuade the accused to confess (Brunk 1979; Yant 1991). Huff and Rattner (1988)
suggest that plea bargaining abuses, especially the piling on of numerous charges by the
prosecutor in order to up the ante of a defendant going to trial, are responsible for many
defendants pleading guilty to a crime they did not commit rather than taking the risk of facing a
128
long prison term (see also Bordenkircher v. Hayes 1978; Gregory, Mowen, and Linder 1978). In
some cases the prosecutor may offer to recommend a reduction in sentence to time-served if a
defendant will confess; in capital cases the prosecutor may offer the defendant an agreement not
to request the death penalty. In sum, prosecutors are often in the position to offer to the innocent
suspect, in exchange for a guilty plea, liberty; in some cases prosecutors offer the suspect an
option of choosing between life and death.
In at least two instances the Supreme Court has heard cases where defendants repudiated
their confessions and sought to have their cases retried because they alleged that their
confessions were not voluntary but were instead offered to avoid the death penalty. In both cases
(Brady v. U.S. [1970] and North Carolina v. Alford [1970]) the Court refused to order new trials.
The 1982 trial of Harry Siegler illustrates the high-stakes game a defendant may be forced to
play as the result of plea-bargaining offers from the prosecution. While Siegler was waiting for a
jury to issue a verdict in his first-degree murder case, he panicked – fearful that he would be
found guilty and be sentenced to death. The prosecutor had offered to drop the death
specification if Siegler plead guilty and, just minutes before the jury returned, Siegler desperately
changed his plea from innocent to guilty of a lesser charge. Siegler then learned that the jury had
already voted to acquit him. Unfortunately for Siegler his guilty plea took precedence and the
judge sentenced him to 60 years in prison. In cases involving the death penalty, Huff, et al.
(1986: 529) note, “with so much to lose, who among us would not plead guilty if he though that
by doing so, he could save his own life, and perhaps, eventually go free when the error is
discovered?”
Adding to the innocent suspects’ predicaments during periods of police and prosecutorial
coercion is the fact that the defendants’ own defense attorney may also believe them to be guilty.
129
In the same way that prosecutor and police authorities are skeptical to claims of innocence,
defense attorneys also usually treat such claims with cynicism (McCloskey 1989). Defense
attorneys may not adequately defend their client’s during periods of coercion because attorneys
may feel pressure from prosecutors and other court personnel to play the game of speedy
disposal of cases (Heumann 1978). Wrongfully accused individuals thus find themselves under
extreme pressure from prosecutors and their own defense attorney to falsely confess or possibly
suffer severe consequences.
Voluntary False Confessions
Bedau and Radelet, in their study of 350 cases of wrongful convictions found a
“disturbing number of cases, where innocent defendants had confessed voluntarily, even eagerly,
but falsely” (1987: 62). One type of voluntary false confession comes from individuals who are
mentally unstable. According to Huff et al. (1996: 112), “some people confess because certain
crimes have caught the public eye; they claim guilt because of their own emotional disturbances,
their need for the limelight, or the fantasies of power they create for themselves.” Huff et al. cite
the California case of the unsolved murder of the young attractive woman who became know as
the ‘Black Dahlia’; her case drew almost 2,000 confessions from individuals of questionable
mental stability. Criminal justice officials, at times, may take advantage of an individual’s
reduced metal capacity, and accept a “confession” that will close a case. In 1983 Ft. Lauderdale
police extracted a false confession to a double murder from mentally handicapped John Purvis –
who was sentenced to life plus two twenty-year terms; after nine years of incarceration Purvis
was released when the actual killers were caught (Leo and Ofshe 1998). In 1979 a mentally
handicapped adult named Melvin Reynolds, after thirteen hours of interrogation, falsely
130
confessed to the abduction and murder of a four-year-old boy. Reynolds was convicted and
sentenced to life in prison. He was released from prison four years later after a serial killer
confessed to a series of brutal murders – including the one for which Reynolds was found guilty
(Leo and Ofshe 1998).
False confessions can be voluntarily tendered in order to deliberately shield someone else
from harm (e.g., a spouse, relative, friend, lover). Sometimes, when a husband and wife are
accused of a crime, the man will falsely confess so the wife can go home with the children. In
other cases a father or mother will falsely confess to protect their child. In 1973 Phoenix,
Arizona, police extracted a false confession from John Knapp who admitted to setting a fire at
his home that killed two of his children. There was no inculpatory evidence supporting the
confession, but considerable evidence supporting Knapp’s innocence. Based on the confession
Knapp was convicted of murder and arson. It was later discovered that one of Knapp’s children
had set the fire, and that Knapp had confessed to protect the child.
False confessions are also voluntarily made in order to protect oneself. For example, in
1944 Delphine Bertrand pleaded guilty to manslaughter and was sentenced to ten to fifteen years
in prison. She falsely confessed to the murder because she was having sex in another part of the
house when the murder took place and, rather than publicly reveal her sex life, she chose to
confess (Yant 1991). In 1995, 62-year-old Laverne Pavlinac falsely confessed to a murder and
implicated her boyfriend in order to escape from their abusive relationship. She later recanted
but was convicted anyway, and she and her boyfriend were sentenced to life in prison (Humes
1999). A few years later the real murderer was caught and provided details of the crime that
only the killer could know (and that Pavlinac had gotten wrong).
131
Lastly, some voluntary false confessions occur simply because a defendant’s legal
expenses are mounting, and he or she cannot afford to continue defending themselves. Rather
than possibly losing their home or a lifetime of savings, defendants may falsely confess to a
crime they did not commit because the probable consequences of the confession are considered
less threatening than the possible consequences of going to trial.
Confessions from Three Innocent Men –
The Cases of Joseph Broderick, George Bigler and Rudolph Sheeler
Philadelphia patrolman James T. Morrow was murdered while tracking down a suspected
robber who had been terrorizing the northeast section of the city. Philadelphia police, in efforts
to solve the murder, arrested and extracted confessions from three different innocent men over a
several year period – two of who were convicted and sentenced to life in prison before being
exonerated.
During the first few months of the investigation following Morrow’s murder, police
arrested Joseph Broderick and quickly extracted a confession from him. A few days later
Broderick recanted. When it became evident to police officials that the confession was coerced,
Broderick was released. Approximately one year later Philadelphia arrested another suspect,
Bigler, for Morrow’s murder. Bigler then became the second man to confess to Morrow’s
murder. In his confession Bigler implicated a Philadelphia patrolman as an accomplice in the
murder. At his trial Bigler repeated his confession and the jury promptly found him guilty of
first-degree murder and recommended that he receive the death penalty. However, the case
against the patrolman Bigler had implicated quickly fell apart and that trial ended in an acquittal.
The judge in Bigler’s trial then became suspicious of the confession and ordered a new trial for
Bigler. At the second trial Bigler again pled guilty and the judge had no alternative but to
132
sentence him; still unsure of the “confession”, the judge sentenced him to life in prison instead of
death. Two years later the same type of robbery that had been attributed to Bigler began to
reoccur in northeast Philadelphia. Police received a tip that the robber was a known criminal
named Jack Howard. When police tracked Howard down, they mortally wounded him in a
gunfight. In Howard’s possession was the murder weapon that had been used to kill Officer
Morrow. Although police had no reason to believe that Howard had an accomplice, they staked
out the hospital room of a friend of his, Elizabeth Morgan, to see if any of Howard’s
acquaintances might visit her. When Morgan’s brother, Rudolph Sheeler, came to visit his sister
he was immediately arrested and taken to police headquarters. Sheeler became the third man to
confess to Officer Morrow’s murder after he was beaten for hours at a time over a two-week
period and finally confessed to aiding Howard in the murder. At trial Sheeler pled guilty and
was sentenced to life in prison. Bigler, who by this time had spent two years in prison was
pardoned and transferred to a mental hospital.
Twelve years passed until proof surfaced that Sheeler was at work hundreds of miles
away at the time of officer Morrow’s murder. A judge reviewed the case and found that key
details of the case were contradicted by his confession, and that his confessions and court
statements contradicted each other. The judge concluded that Sheeler had been forced to confess
because police were eager to free Bigler and therefore clear the reputation of the officer he has
implicated – even though that officer had been acquitted. The Pennsylvania Supreme Court,
calling the case “a black and shameful page in the history of the Philadelphia police department”
(Yant 1991: 49), overturned Sheeler’s conviction and ordered his immediate release. Four
detectives and two superior officers were suspended for their roles in Sheeler’s coerced
confession.
133
Extent Of the Problem of False Confessions
False confessions occur frequently (see Miller v. Illinois, 1977; Blackburn v. Alabama
1960; Bumper v. North Carolina 1968;Chambes v. Florida 1940). Bedau and Radelet (1987:
58), in their study of 350 cases of wrongful convictions in capital, or potentially capital cases,
found that in 14 percent of the cases they studied, police coerced innocent defendants into
confessing to crimes they did not commit by using methods that “ranged from those in which
the police used the ‘third-degree,’ to others where less brutal tactics were employed.” Leo and
Ofshe, in their 1998 study of sixty cases of police-induced false confessions in the post-Miranda
era, found that American police continue to elicit false confessions even though the era of third
degree interrogation has purportedly passed. Scheck, et al (2000) found that “false confessions”
were a factor in 24 percent of their cases, and rated it seventh among the most prevalent factors
leading to wrongful conviction.
Conclusion
False confessions are one of the least publicized aspects of wrongful conviction. A
probable reason why so few instances come to the public’s attention is because the individuals
are often given probation, suspended sentences, or sentences limited to time-served. There is,
therefore, usually little incentive for anyone to pursue the case further. Police officials and
prosecutors typically apply pressure on a suspect to confess because they believe that the suspect
is, in fact, guilty, and they wish to speedily dispose of the case. Defense attorneys, who may
themselves be cynical of their client’s claim of innocence, may not vigorously defend their
client’s interests, but instead may respond to pressures from the courtroom workgroup to close
134
cases. Wrongfully accused individuals thus find themselves under extreme pressure to falsely
confess or possibly suffer severe consequences.
False confessions are problematic to the criminal justice system. Leo and Ofshe (1998:
432) explain,
A lack of information prevents researchers from
estimating the full magnitude of personal and social
harm that police-induced false confessions cause: the
days and months innocent persons spend in pre-trial
incarceration; the resources, time, and dollars wasted
prosecuting and defending them; the months and years
defendants languish in prison after wrongful conviction;
and the additional crimes carried out by the true
perpetrators.
Huff et al (1996) suggest that better police training can reduce the incidence of false
confessions, with an emphasis on the investigators’ duty to uncover the truth – and not to break
big cases whatever the cost to further their own careers. Prosecutors, they suggest, can reduce
the incidence of false confessions by more carefully scrutinizing the details and discrepancies
contained within confessions, and by considering the motives or conditions that might lead to a
confession. Also confessions that are repudiated, and which have no other collaboration, should
be deemed insufficient to warrant a conviction.
Leo and Ofshe (1998) suggest that if police and prosecutors wish to prevent false
confessions, “they must acknowledge the reality of false confessions, seek to understand their
causes and consequences, and work to implement policies that will both decrease the likelihood
of eliciting false confessions and increase the likelihood of detecting them (1998: 492). They
further suggest that shoddy police practice often results from poor interrogation training, and the
use of influential training manuals that have been shown to be coercive and produce false
135
confessions. Leo and Ofshe suggest that the incidence of false confessions could be reduced if
legislatures would mandate that all interrogations be recorded in their entirety.
136
CHAPTER TEN
WRONGFUL CONVICTION AND COMMUNITY PRESSURE
Behind [the mob] is the cry for blood, but it is muted, more gentle in appearance, and
sugar-coated with law-abiding respectability. It manifests itself in various ways: people
are emotionally aroused by a crime and display above-normal interest; there is a restless
rumbling by this disturbed public and a sharp increase in letters and telephone calls to
the police, to newspapers, to officials, all insistently demanding action. Newspapers
reflect the general temper of its readers and become more shrill, and so the heat is on.
Police, prosecutors, and even judges may be goaded into hasty action; judgment becomes
blurred under this pressure; suppositions replace facts, and since jurors are selected
from among the excited public, little attention is paid to actual evidence, the guilty
verdict is a foregone conclusion, and the net result is a legal lynching. And a study of
cases where innocent persons have been convicted shows that this happens more
frequently than is realized. Edward D. Radin “The Innocents” (1964: 55)
Pressure from the public to convict a criminal suspect can be viewed as an expression of
democratic participation in the criminal justice process. Public pressure, therefore, can make the
criminal justice system more responsive to the needs of a community. A question of interest is
whether society is willing to tolerate wrongful conviction in exchange for an increased sense of
deterrence and public safety. Huff and Rattner describe community pressure for conviction as a
time “when hatred replaces reason and due process” (1988: 158). Public pressure exerted to
solve a crime, sometimes fanned by newspaper publicity, is often as much to blame for a
wrongful conviction as is eyewitness error, faulty science, or official misconduct. Police,
prosecutors, judge and jurors are not impervious to public demands for revenge and closure in
high profile criminal trials that take place in highly charged settings where public interest and
passions are aroused. Huff et al. (1996) suggest that in periods of high crime rates and great
public outcry against criminals, and when group pressures are felt in the courtroom, conviction
rates may be higher than at other times. They note, “it is undeniable that efforts to influence
what is happening in the world often extend to the courtroom” (1996: 75). A review of the
137
literature on wrongful conviction suggests that when community pressures for solving a crime
increases, there is a tendency for carelessness and lawlessness to also increase, and that the
probability of a wrongful conviction increases when enforcement agencies are willing to respond
to the public’s demand for punishment (Rattner 1983). Humes (1999) suggests that a growing
national trend is occurring in which innocence defendants have become unintended causalities of
the current war on crime.
Wrongful convictions that result from community pressure have been a major concern
of many researchers. Borchard (1932), in his pioneering study of wrongful conviction, noted
that “community opinion demanding a conviction” contributed to 22 percent of the 65 cases of
wrongful convictions in his study. Radin (1964) devoted 30 pages of his book on wrongful
conviction to the topic of community pressure. Bedau and Radelet (1987: 63) noted that,
“sometimes community outrage over a crime turns the criminal proceedings into a near lynching,
as the trial degenerates into a kangaroo court;” they determined that “community opinion
demanding a conviction” contributed to 22 percent of the erroneous convictions in their study.
The Case of Clarence Brandley
On August 23, 1980, Brandley was a janitor in a high school in Belleville, Texas where a
girls’ volleyball tournament was being played. While the team was warming-up someone
noticed that the female student-manager was missing. An exhaustive search was made of the
high school until Brandley and another janitor, Henry ‘Icky’ Peace, found the 16-year-old blondhaired
girl’s body, strangled and nude, in a loft where theater props were kept. She had been
raped.
138
With a rapist and murderer loose in the community, townspeople panicked. Radelet et al.
(1992: 120) note, “The town was in an uproar. Summer was coming to an end, and the school
year was about to start. Parents threatened to keep their children (especially the girls) home
unless the murderer was arrested by the time school registration was completed. Town and
school officials alike recognized that they had to act fast if the growing panic and outrage was
not to overwhelm the community.” Suspicion quickly fell on the two janitors who discovered
the girl’s body. Two days after the crime they were both given a lie-detector test, but both men
passed and were released. Four more days passed and time was running out for authorities
because school was to begin the following Monday. Intense meetings were held between the
police and the county district attorney to find a way to quickly solve the case so that school could
begin on time. It was decided to arrest Clarence Brandley for capital murder.
From the beginning, Brandley insisted he was innocent. Five days after his arrest he
voluntarily appeared before an all-white grand jury. He was indicted and then tried by an allwhite
petit jury. In Brandley’s first trial one juror refused to vote for guilt, and the jury was
hung. When the identity of the dissenting juror was publicly revealed, he received numerous
harassing telephone calls and threats from local citizens. At Brandley’s second trial, also before
an all-white jury, he was convicted of murder and sentenced to death. Eleven months after
Brandley’s conviction his appellate attorney discovered that more than half of the evidence used
to convict Brandley had mysteriously disappeared – including the rape kit containing the killer’s
sperm, and three reddish-blond hair samples of Caucasian origin that were found on the victims
nude body. An appellate court was unsympathetic with the claim that Brandley was denied a fair
trial because the evidence that could exonerate him was lost or purposely destroyed by criminal
justice authorities. Brandley was given his first execution date: January 16, 1986.
139
The first break in Brandley’s case came when the wife of ex-janitor James Robinson told
investigators that her husband had admitted to killing the girl. Friends of Robinson’s wife urged
her to contact the state district attorney, but when she did the district attorney did not believe her
story. An investigation revealed that James Robinson was at the high school visiting with
another janitor, Gary Acreman, on the morning of the murder. When Acreman’s father-in-law
was questioned, he told investigators that his son-in-law had told him where the victim’s clothes
had been hidden – two days before police were able to locate them. Finally another janitor
admitted to investigators that he had seen Robinson and Acreman grab the screaming girl while
she was in the woman’s restroom.
Brandley’s attorney filed a writ of habeas corpus to present the new evidence. A judge
issued a stay of execution on March 20, 1987 – six days before Brandley’s second execution
date. At the evidentiary hearing the judge found that the State’s investigation of the crime was
“so impermissibly suggestive that false testimony was created, thereby denying Brandley due
process of law and a fundamentally fair trial ” (Radelet et al 1992: 133). In December 1989, the
judge ordered a new trial; the prosecutor, however, refused to agree that Brandley could be
released on bond pending the new trial. Brandley remained in prison for almost another year
until October 1, 1990, when, after more than ten years of incarceration on death-row, all charges
were dismissed against Brandley and he was released from prison.
Reducing the Effects of Community Pressure
Community pressure for conviction is not limited to instances of racism. Pressure may
also be exerted by woman’s groups, homosexual and anti-homosexual interests, child-protection
140
interests, and even business groups who want crimes solved because business interests are being
adversely affected. Sometimes these pressures are exerted not only by local citizens, but also by
national special interest groups. Recent attention to domestic violence (Taylor, Davis and
Maxwell 2001), date-rape (Kingsworth, MacIntosh, and Wentworth 1999), and child molestation
(Humes 1999) are reflections of increased national and community sensitivity to these issues,
and thus increased community pressure on criminal justice officials to obtain convictions for
these types of crimes. Huff et al. (1996) suggest that the increasing number of instances of
wrongful imprisonment in rape cases since the 1970’s might be a result of increasing public
pressure to obtain convictions in such cases. Another type of community pressure can be fanned
by police organizations who seek to compel prosecutors, judges, and jurors to convict accused
“cop-killers” – as was demonstrated in the wrongful conviction cases of Randall Adams, Joseph
Broderick, George Bigler, and Randolph Sheeler.
In some cases community pressure for conviction can be mitigated with a change of
venue. If mob violence threatens a case, a change of venue must be allowed (Blevins v. State,
1963); if an impartial jury cannot be selected in the original venue, a change of venue must be
granted (People v, Harris, 1981). In State v. Engle (1980) the Court determined that, in order to
grant a fair trial, the court must consider all relevant factors: e.g., the nature and gravity of the
offense, the amount of pre-trial publicity, the size of the community, the defendants status in the
community, the prominence of the victim, and the likelihood of finding a substantially fairer jury
elsewhere. Even in cases where venue is changed, however, there is no guarantee that trials will
be free of community pressures. Newspapers in the newly venued community may pick up the
story and sensationalize it in a way similar to the way the story was sensationalized in the
community of original jurisdiction. Issues of the case may appear so outrageous that citizens in
141
‘any’ community might feel the need to pressure criminal justice authorities to convict the
accused.
Conclusion
Numerous cases presented in this dissertation have demonstrated the effect of community
pressure on wrongful conviction. In the case of James Richardson who was wrongfully
convicted of killing his seven children, Radelet et al. (1992: 197) noted, “News of the tragedy
that struck the dusty Florida town of Arcadia in October, 1967, spread quickly. Reporters
flocked to the region. For a few days, the deaths of those seven children, assumed to be
murdered, was one of the top stories in the nation . . . . . The national media was on hand, and the
pressure on [sheriff] Cline to make an arrest was strong.” Twenty years after the news-blitz, the
innocent Richardson was exonerated and set free from prison. The high-profile case of Dr. Sam
Shepard, convicted in Ohio in 1954 of the brutal murder of his pregnant wife, was described as
“a circus, with the judge as ringmaster” that took place in a “cyclone of publicity” (Bedau and
Radelet 1987: 32). After serving several years in prison Shepard’s innocence was established,
his conviction was set aside, and he was released from prison.
Public pressure to convict is a double-edged sword that can cut two ways: it can serve as
a watch dog, forcing prosecution of cases that might otherwise be dropped because of an
accused’s standing or influence in the community, or it can be used as a tool of fear and
prejudice by community members to persecute and prosecute, as scapegoats, wrongfully accused
defendants who have few or no resources. Community pressure for conviction also has an
antithesis – community apathy. Community apathy is undoubtedly more destructive to the cause
of the wrongfully accused than is community concern. During abnormal periods of community
tension which may arise as the result of a high-profile crime, at least the community is
142
“involved” in the criminal justice process; i.e., citizens are actively exerting pressure on
authorities to see that justice is served.
143
CHAPTER ELEVEN
WRONGFUL CONVICTION AND “OTHER” FACTORS
Previous sections of this dissertation explored research concerning the eight factors most
associated with wrongful conviction. Also deserving mention are the following, less empirically
studied, but also important “other” factors associated with wrongful conviction.
Presumption of Guilt – A cornerstone of the U.S. criminal legal system is the
presumption of innocence. Yant notes, “ Despite all the alleged protections built into the trial
process, the sad fact is that a person facing criminal charges is actually presumed guilty until
proven innocent rather than presumed innocent until proven guilty” (1991: 11). The
‘presumption of guilt’ attitude held by many criminal justice officials adversely affects
wrongfully accused individuals at every step of the criminal justice process (Lane 1970; Linscott
1994; McCloskey 1989; Packer 1968; Rattner 1983; Scheck et al. 2000). James McCloskey
(1989) lists ‘presumption of guilt’ as the number one factor leading to wrongful conviction. He
states that most people are inclined to believe the adage where-there-is-smoke-there-is-fire; laypeople
and criminal justice professionals alike tend to believe that suspects probably committed
the crime of which they are accused.
Klockars suggests that police officers “tend to assume guilt as a working premise of their
craft” (1980: 37), and are inclined to use a “not guilty (this time) assumption,” during stops with
seemingly innocent citizens. In these instances police officers may be predisposed to believe,
even if no evidence of unlawful behavior is initially found, that a suspect may be
hiding an undisclosed truth regarding a previous crime, and that a little additional pressure might
cause it to be revealed.
144
Existence of a Prior Criminal Record – is a key factor in the wrongful conviction of innocent
individuals (Huff et. al 1996; McCloskey 1989; Rattner 1983, Yant 1991). Borchard (1932), in
his pioneering study of wrongful conviction, determined that “prior convictions and ‘unsavory’
record” contributed to 35 percent of the convictions he studied. Once judges or jurors know that
a defendant has a prior criminal record, they tend to label the suspect as a “criminal” and any
evidence presented against him or her will be considered from that perspective. Huff et al.
(1996) note that a prior criminal record may increase an individual’s chances of being wrongly
accused because: (a) it places that individual’s picture in police files (that victims of crimes are
constantly viewing), and (b) it makes an individual more venerable to police interrogation and to
making false confessions. Innocent defendants with criminal records often find themselves in
no-win situations when their cases go to trial. They are less likely to take the witness-stand and
testify in their own behalf because their record might be questioned during cross-examination,
and the refusal of a defendant to take the witness-stand is often viewed by jurors as a sign of
guilt (McCloskey 1989; Rattner 1983).
Race – Just as our prisons contain a disproportionate number of blacks and Hispanics, so are
there a disproportionate number of convicted innocents whom are black or Hispanic (Huff et al
1996). The biases and prejudices of police, prosecutors, judges, and jurors can be a factor in any
case of wrongful conviction. Prejudicial justice is common in most states, and many wrongful
convictions are the result of racial motivations (Yant 1991). Scheck et al. (2000) note that
racism and bigotry, written out of the law, still shadows police precincts, courtrooms and jury
boxes. Summarizing their findings concerning wrongful conviction, they state, “among the more
troubling findings, is that several of these factors (eyewitness misidentification, false
145
accusations, false confessions, police and prosecutorial misconduct, inadequacy of counsel) are
more pronounced in the conviction of innocent black men” (2000: 246; see also Huff et al.
1996).
Racism is even found in the areas or forensic science (Castelle 1999; Scheck et al. 2000).
Documents presented in a 1997 Inspector General’s investigation of the FBI Crime Lab contain
disturbing allegations of fraud and racism. According to one interview contained in the report,
FBI agent Greg Parsons stated that a lab technician “would determine if the suspects were Afro-
Americans; if so, he would manipulate the tests to prove guilt” (Office of Inspector General,
Memorandum of Investigation, 1/23/97).
Poverty – Clarence Darrow is credited with making the cynical observation, “I’ve never heard of
a millionaire going to the chair” (Radin 1964: 233). In a majority of cases of wrongful
conviction, the defendants had one thing in common – a lack of money. Poverty will first impact
a defendant when bail is set. When indigent defendants are sitting in jail there is a extreme
pressure on them, even when innocent, to plead guilty to a crime they did not commit in
exchange for a recommended reduced sentence. Even when a wrongfully accused defendant has
the resources necessary to hire his own attorney, he still may lack sufficient funds to make a
thorough investigation of the cases against him; attorney fees may be small in comparison to the
funds needed to hire investigating detectives, or to hire technical experts to examine and testify
regarding physical evidence. The prosecution, with relatively unlimited investigative resources,
can overwhelm all but the richest of defendants.
146
Mental Incompetence – is a factor in wrongful conviction when the accused is intellectually
unable to articulate a defense, or does not have enough mental capability to be considered legally
culpable of committing a crime (see Ake v. Oklahoma 1985; Huff et al. 1986). Rattner (1983)
reports several known cases of false convictions that involve persons of low I.Q.; he notes that
mentally incompetent defendants are often unable to inform their attorneys of salient facts which
might exonerate them, and were often unable to make a satisfactory impression on the jurors.
Radelet et al. (1992), in their study of erroneous convictions in capital cases, discuss almost twodozen
separate cases where individuals with low I.Q.’s were wrongfully convicted of capital
crimes.
Jurors – The jury system used in the United States is not perfect, but it may represent the best
system available for meting out justice. The general consensus among criminal justice
authorities is that juries are right more often than they are wrong (McCloskey 1989). The idea
that juries are right more often than they are wrong, however, presents small comfort to the
innocent defendant convicted by one of the ‘wrong’ juries. Jurors, according to Scheck et al.
(2000: 109), are “panning for nuggets of truth in the muddy rivers of conflicting stories and
rickety memories.” They are ultimately forced to make one subjective call after another when
deciding whom to believe and what inferences to draw from conflicting statements.
Radin (1964: 230), however, exonerates juries from most of the blame regarding
wrongful convictions,
In my research I have found that the jury is seldom a primary cause
of the conviction of the innocent. To blame jurors would be an
easy way to find a scapegoat. A jury has every right to believe that
police and a prosecutor have checked thoroughly into the
identification by an eyewitness, whereas this usually marks the end
of any further investigation by authorities. If the judge allows the
147
courtroom to become a theater for histrionic displays by
contending defense counsel and the district attorney, instead of a
calm, impartial forum where facts are presented and truth is
sought, the fault is more with the judge than the jury if jurors are
allowed to be carried away by emotions.
Judges – Responsibility for allowing all types of misleading evidence to enter the courtroom
ultimately falls on the bench. Rattner (1983) lists three major types of judicial error associated
with wrongful conviction: judicial errors affected by bias, judicial neglect of duty, and technical
errors in judicial decisions. Often wrongful convictions occur because judges allow incompetent
defense attorneys, overzealous police and prosecutors, and questionable forensic evidence to
permeate the courtroom. Sometimes the biases of the judges themselves lead to trial errors that
are not reversible but never-the-less influence decision-making (Huff et al. 1996; Rattner 1983).
Sheck et al. (2000) note that the occurrences of wrongful conviction are attenuated when
appellate judges issue only one-line judicial orders that do not specify the reasons for overturning
a conviction.
Simple Mistake – wrongful conviction of the innocent can occur even when the criminal justice
process is operating as it should. Radin (1964) suggest that a “sardonic fate” sometimes seems
to operate in some cases of wrongful conviction. These are the types of cases in which
eyewitnesses honestly make mistakes, police departments use proper investigative techniques,
prosecutors are zealous but not over-zealous, defense attorneys do a credible job of defending
their client, and still an innocent person is still convicted. No system is perfect. A system in
which honest mistakes occur can be tolerated – if those mistakes are minimized; a system in
which preventable errors are continually made and repeated should not be tolerated.
148
CHAPTER TWELVE
RESEARCH METHODOLOGY
Introduction
This chapter describes the methods used in this study. The study’s objectives were to
obtain from criminal justice professionals their perceptions regarding: (1) the frequency of
wrongful conviction in their own jurisdictions, and in the United States, (2) the frequency of
occurrence of factors which previous research has associated with the incidence of wrongful
conviction, and (3) the general reliability of eyewitness and forensic expert testimony and the
general reliability of the work product of police officials, prosecutors, defense attorneys, and
judges.
In order to examine these issues a 53-item survey questionnaire was administered to four
groups of Ohio criminal justice professionals: law enforcement (sheriffs and chiefs of police),
prosecutors (chief and assistant), defense attorneys (private and public defenders), and judges
(Common Pleas, Appellate, and Supreme Court). The opinions of these professionals were
sought for two reasons: (1) they are close to the trial process and are in a position to observe
incidents of wrongful conviction, (2) their experience should provide them with knowledge of
criminal justice system successes and failures
State of Ohio criminal justice professionals were targeted for this study because the use
of a single large state, such as Ohio, serves to control for the effect of varying legal definitions,
while still allowing for a diversity of settings. Ohio is the sixth largest state in the United States,
with a population of 11,353,140 (Bureau of Census, 2001), and has a blend of urban, suburban,
and rural areas.
149
The “perceptions” of criminal justice professionals were chosen as a measure because
there may be no better way to approach the question of prevalence than through perceptions.
Due to the fact that no research approach has yet been devised to exactly measure the extent of
the problem, the perceptions elicited in the present study may offer a ‘best-estimate’ of the
phenomenon. The professionals surveyed are in a position to make such observations because
they are exposed, on almost a daily basis, to the environment where wrongful convictions can
occur. Due to their experience within the criminal justice system, their perceptions regarding
the issues of this study may enable them to offer data close enough to objective truth to provide a
baseline estimate of the frequency of system errors.
Section one of this chapter discusses the research questions that guide the study.
Section
two provides a discussion of the sampling frame and sample size. Section three focuses on data
sources. Section four discusses survey administration/data collection. Section five discusses
variable measures. Section six outlines the statistical analyses used to interpret the data, and
provides a description and rationale for these techniques. Finally, section seven discusses the
limitations of the study.
Research Questions
The research reported here is designed to answer the following questions regarding the
perceptions of criminal justice professionals:
1. How frequently does wrongful felony conviction occur?
2. How frequently do criminal justice professions (police officials, prosecutors,
defense attorneys, and judges) commit errors associated with the incidence of
wrongful conviction?
150
3. What is the general reliability of the work product of police officials, prosecutors,
defense attorneys, and judges?
4. How frequently do non-criminal justice personnel commit errors associated with the
incidence of wrongful conviction (eyewitness error, forensic witness error, false
accusations, false confessions, and community pressure)?
5. What is the general reliability of eyewitness and forensic expert testimony?
6. What do criminal justice professionals consider an “acceptable” level of wrongful
convictions, and do they believe changes warranted in the criminal justice system?
7. Do groups of criminal justice professionals differ in their perceptions as they relate
to the issues delineated in the first six research questions?
Sampling Frame and Sample Size
Sampling Frame
The sampling frame consists of: (1) all Ohio Common Pleas Court judges, (2) all Ohio
Appellate Court judges, (3) all Ohio Supreme Court judges, (4) all Ohio chief county
prosecutors, (5) a random sample of Ohio assistant prosecutors, (6) all Ohio county sheriffs, (7) a
random sample of Ohio chiefs of police, (8) a random sample of Ohio county public defenders,
and (9) a random sample of Ohio private defense attorneys.
Sample Size
The sample size consists of: (1) 230 presiding Common Pleas Court Judges, (2) 67
Appellate Court Judges, (3) 7 Supreme Court judges, (4) 88 chief county prosecutors, (5) 132
assistant prosecutors, (6) 88 county sheriffs, (7) 400 chiefs of police, (8) 250 county public
151
defenders, and (9) 238 private defense attorneys. The total sample comprises 1500 Ohio
criminal justice professionals.
Sample sizes were determined using Dillman’s (2000) “Tailored Design Method for Mail
and Internet Surveys.” The Dillman (2000) method utilizes a formula based on: the amount of
sampling error that can be tolerated, the population size from which the sample is to be drawn,
how varied the population is with respect to the characteristics of interest, and the required
confidence level. Sample sizes for the present survey were determined using the formula to
target a 95% confidence level, with a +/- 5 percent sampling error.
Date Sources
Ohio Common Pleas Court, Appellate Court, and Supreme Court Judges
Names of members of the Ohio Common Pleas Judges Association were obtained
from the Association’s 2001 President, Hon. Margaret K. Weaver. Judge Weaver provided a list
of all 230 Ohio Common Pleas Court judges, all 67 Ohio Appellate Court judges, and all 7 Ohio
Supreme Court judges. Judge Weaver was succeeded by the association’s 2002 president, Hon.
James R. Williams, who approved a letter stating the association’s cooperation with the survey.
All 304 judges were mailed questionnaires.
Ohio Criminal Defense Attorneys
A list containing the names of approximately 23,000 attorneys was obtained from the
Ohio Bar Association 2000 “Official Legal Directory.” The Association’s web page was used to
identify Association members who specialize in private practice criminal law. A random
numbers table was used to randomly select 238 private criminal defense attorneys who were
152
mailed questionnaires. The Association’s website also contains a list of Ohio public defenders.
Again, a random numbers table was used to randomly select 250 Ohio public defenders who
were mailed questionnaires.
Ohio Prosecutors
A list of names of all 88 Ohio chief county prosecutors was provided by Mr. John
Murphy, Executive Director of the Ohio Prosecuting Attorneys Association. All Chief County
Prosecutors in Ohio were mailed a questionnaire. The names of Assistant County Prosecutors
were obtained from the Ohio State Bar Association’s website. A random numbers table was
used to randomly select the 132 Assistant Ohio Prosecutors who were mailed questionnaires.
Ohio County Sheriffs
Names of all 88 Ohio county sheriffs were obtained from a list acquired from the Ohio
Attorney General’s “2001 Law Enforcement Directory.” All 88 county sheriffs were mailed a
questionnaire.
Ohio Chiefs of Police
The names Ohio’s 857 chiefs of police were obtained from a list acquired from the Ohio
Attorney General’s “2001 Law Enforcement Directory.” A random numbers table was used to
randomly select 400 chiefs who were mailed questionnaires.
Survey Administration/Data Collection
In order to minimize the four sources of survey error (sampling error, coverage error,
measurement error, and non-response error), the survey was administered using a version of
Dillman’s (2000) Tailored Design Method (TDM). As recommended by Dillman, multiple
153
mailings were conducted to improve survey response rates. (The survey instrument and
accompanying letters are included in Appendix A). Initially, each member of the sample was
mailed a “pre-notice” postcard introducing the study, announcing that a survey would soon
follow, and emphasizing the importance of each respondent to the study. Three days later, each
member of the sample was mailed a questionnaire, an addressed postage-paid return envelope,
and a cover letter which introduced the survey and reemphasized the importance of each
individual’s response. The cover letter was limited to one page and briefly asked for the
respondent’s help, stated why the survey was important, and assured confidentiality to
respondents. Two weeks after the original survey was mailed, a follow-up “thank you -
reminder” postcard was sent to all non-respondents to remind them to return a completed survey
(or to thank them if the survey had recently been returned). Two weeks after the “reminder”
postcard was mailed, a second survey, cover letter, and addressed postage-paid return envelope
was mailed to all non-respondents – again reminding them of the importance of their response
and urging them to fill out and return the questionnaire. Four weeks after the second
questionnaire was mailed, a final questionnaire, cover letter, and addressed postage-paid return
envelope was mailed to the remaining non-respondents with a final plea for a response. This
reminder-letter was insistent in nature – simply because it constituted a fifth request. It again
reminded the respondent of the importance of the study as well as the importance of the
respondent’s participation. Using Dillman’s five-step TDM method, a response rate of 53.2
percent was attained.
154
Study Variables And Measures
Measures/Variables
In general, previous research concerning the topic of wrongful conviction discussed the
following issues: frequency of wrongful conviction, eyewitness error, error concerning the
presentation of forensic evidence, law enforcement error, prosecutorial error, inadequacy of
defense counsel, judicial error, false accusations and false confessions, community pressure to
obtain a conviction, and presumption of guilt. The survey instrument (Appendix A) contains 53
questions intended to tap the perceptions of respondents concerning these issues.
Frequency of Wrongful Conviction. Two questions using 10-item scales ranging from
‘zero percent’ to ‘over 25 percent’ were utilized to allow respondents to estimate what they
perceive to be the percentage of wrongful felony conviction occurring in their own jurisdiction,
and in the United States in general. A third question, using the same 10-item scale, allowed
respondents to indicate what they consider to be an “acceptable level” of wrongful convictions.
Professional and Non-Professional Error. Following is a description of how the
variables which are the subject of this study are operationalized. (All “9-item scales” used in the
survey instrument ask respondents to estimate, in a range from “never” to “always,” how often
they perceive a specific type of error to occur.)
Eyewitness Error. Three questions were utilized to operationalize this variable. Two
questions asked respondents to estimate how often (never, seldom, often, very often) they believe
eyewitnesses misidentify a defendant – either intentionally or in good-faith. A third question
asked respondents to estimate how reliable (very reliable, usually reliable, usually unreliable,
very unreliable) they believe are the testimonies of eyewitnesses in criminal cases.
155
Forensic Expert Error. The survey contained two 9-item scale questions asking
respondents how often they believe forensic experts – either in good-faith, or intentionally –
misrepresent forensic evidence. A third question asked respondents to estimate how reliable
(very reliable, usually reliable, usually unreliable, very unreliable) they believe are the scientific
conclusions of forensic experts.
Police Error. Three questions were used to operationalize this variable. The first
questions, containing five parts, asked respondents to estimate on a 9-item scale how often they
believe police officials are involved in the following five types of police error:
Inadequate police investigation
Police coaching witnesses in pretrial identification procedures
Police suppressing exculpatory evidence
Police using false evidence
Police using undue pressure to obtain a confession
A second question asked respondents to estimate how reliable (very reliable, usually
reliable, usually unreliable, very unreliable) they believe is the evidence presented in court by
police officials. Finally, a question asked respondents how often, in general, they believe police
identification procedures contribute to misidentification (never, seldom, often, very often).
Prosecutorial Error. One five-part question asked respondents to estimate, using a 9-
item scale, how often they believe prosecutors are involved in the following five types of
prosecutorial error:
Inadequate investigation of case by prosecutor
Prosecutor suppressing exculpatory evidence
Prosecutor using undue plea bargaining pressure
Prosecutor prompting witnesses
Prosecutor knowingly using false testimony
156
A second question asked a respondent how reliable (very reliable, usually reliable,
usually unreliable, very unreliable) they believe is the evidence presented in court by
prosecutors.
Defense Attorney Error. In order to operationalize the variable “defense attorney error,”
two questions were asked. The first question again contained five-parts. Respondents were
asked to indicate, on a 9-item scale, how often they believe defense attorneys are involved in the
following five types of defense error:
Inadequate investigation of case by defense attorney
Defense attorney failing to file proper motions
Defense attorney not adequately challenging forensic evidence
Defense attorney not adequately challenging witnesses
Defense attorney making unwarranted plea bargain concessions
Respondents were then asked to estimate how well (very well, moderately well, not very
well, poorly) they believe defense attorneys defend their clients.
Judicial Error. The variable “judicial error” was operationalized by first utilizing a fourpart
question asking respondents to indicate how often they believe judges are involved in the
following four types of judicial error:
Judicial error concerning the admissibility of physical evidence
Judicial error concerning the admissibility of eyewitness testimony
Judicial error concerning the admissibility of expert testimony
Error resulting from judicial bias
A second question sought information regarding how reliable (very reliable, usually
reliable, usually unreliable, very unreliable) respondents believe are the decisions made by
judges concerning defendants’ guilt or innocence.
False-Witness Errors. Respondents were asked how often they believe false accusations
occur: frequently, occasionally, or seldom.
157
False Confessions. This measure is operationalized using a “yes-no” format asking
respondents if they personally knew of a case in which an innocent individual confessed to a
crime he or she did not commit.
Community Pressure. A “yes-no” question was again used asking respondents if they
personally knew of a case in where community pressure to quickly solve a crime lead to a
wrongful conviction.
Sample Demographics. Five questions ask for basic demographic information
concerning respondent’s age, gender, race, professional background, and current jurisdiction:
Statistical Analysis
It was the intention of this dissertation to examine the perceptions of individual actors
within the criminal justice system concerning systemic factors associated with wrongful
conviction, and to compare these perceptions across groups. The main emphasis of the statistical
analysis was to measure the total response rate of all participants regarding each of the study’s
research questions, and to utilize statistical tests to measure variations in frequency of responses
across groups of participants. Following is a description of the types of statistical analysis
techniques and procedures used to analyze the data – and the rationale for using each technique.
The responses to three questions concerning the frequency of wrongful conviction
utilized a ratio scale that allow respondents to select from a range of “zero” to “over twenty-five
percent,” how often they believe wrongful convict does, or should, occur. In order to determine
if differences in means existed in the answers to these questions, an analysis of variance test was
used.
Twenty-two questions utilized an interval scale to allow respondents to select, from the
category of interest, responses ranging from “1” (never) through “9” (always) – with equal
158
intervals between the numbers one through nine. Although this type of scale has some “ordinal”
qualities (order and direction), it also possesses “interval” qualities (equal distances between
observation points). Analysis of variance tests were also performed on this data – a method
currently accepted and in use in the majority of scientific journals. If analysis of variance tests
determined that differences existed among means, post hoc range tests and pairwise multiple
comparisons were used to determine which means differ. Where analysis of variance indicated
significant differences in mean responses across groups, a post hoc Tukey HSD (honesty
significant difference) test was utilized to determine which means differed significantly.
Ten survey questions required responses using an ordinal scale. For these responses only
frequencies are reported. A chi-square test was not able to be used because the data did not meet
the statistical assumptions necessary to run chi-square.
Thirteen questions required nominal-scale responses. For these questions, only frequency
distributions are reported. Finally, one open-ended question, asking respondents for their
opinions regarding what steps can be taken by criminal justice professionals to reduce the
incidence of wrongful conviction, was located at the conclusion of the survey instrument. A
representative sample of these responses is presented in the discussion section in chapter
fourteen.
Limitations of the Study
Although the findings presented in this study advance the current state of knowledge
regarding wrongful conviction, the reader should be aware of limitations in the data. The present
study utilized survey research to ask criminal justice professionals their opinions regarding the
extent and causes of wrongful conviction. In regard to the issue of the “extent” of wrongful
conviction, a limitation of the study is that the actual rate of error was not measured, but only a
159
“perceived” rate of error; the actual extent of wrongful conviction remains an empirical question.
Another limitation of the study involves the “types” of errors which were offered to respondents
for comment. Prior research has indicated that these errors are only “associated” with wrongful
conviction, and have not been proven to be “causative” of wrongful conviction. It is possible,
therefore, that the reported frequency of a particular type of error may be spurious and irrelevant
to the incidence of wrongful conviction. Also, data from the present study was obtained from
only one state, Ohio, and therefore has limited generalizability.
160
CHAPTER THIRTEEN
FINDINGS
Introduction
This chapter is organized to correspond to the research goals outlined in the methods section.
First the survey response rates and the descriptive properties of the response sample
are presented. Then each of the seven research questions are restated. The pertinent survey questions
are listed under each research question, and the findings are presented.
Survey Response Rates
A survey questionnaire concerning the topic of wrongful conviction was mailed to 1500
Ohio criminal justice professionals. A total of 798 (53.2%) questionnaires were completed and
returned. The findings presented in this chapter reflect an analysis of these completed
questionnaires. Table 1 reports the response rates by group.
Table 1 – Survey Response Rates
Group Surveys Mailed Responses Response Rate
Police 488 274 56.1
Sheriffs 88 62 70.4
Chiefs of Police 400 212 53.0
Prosecutors 220 103 46.8
Chief Prosecutors 88 62 70.4
Assistant Prosecutors 132 41 31.1
Defense Attorneys 488 235 48.2
Private 238 98 41.1
Public Defenders 250 137 54.8
Judges 304 186 61.2
Common Pleas 230 142 61.7
Appellate 67 41 61.2
Supreme Court 7 3 42.9
TOTAL 1500 798 53.2
161
Descriptive Properties of the Sample
Of the 798 individuals who responded to the survey, 274 (34.3%) work in law
enforcement (“police”), 103 (12.9%) are prosecutors, 235 (29.4%) are defense attorneys, and 186
(23.3%) are judges. Many respondents have, at some time in their career, gained experience in
more than one of these areas. A comparison of the descriptive variables of the four groups is
reported in Table 2.
Table 2 – Group Comparisons: Sociodemographic Variables
Variables Police Prosecutors Attorneys Judges Total*
Age
18-29 5 7 13 0 25
30-39 32 22 71 4 129
40-49 106 34 66 30 236
50-59 112 37 67 82 298
60-69 18 0 16 50 94
70-79 2 0 4 9 15
80+ 0 0 0 0 0
TOTAL 275 100 237 175 787
Race
White 272 98 219 167 756
Black 0 1 8 2 11
Hispanic 0 0 2 1 3
Asian 0 0 1 0 1
Other 1 1 3 1 6
TOTAL 273 100 234 171 778
Gender
Male 271 86 194 148 699
Female 4 14 43 31 92
TOTAL 275 100 237 179 791
Jurisdiction
Village 98 0 2 0 100
Township 28 2 1 2 33
City 82 2 29 2 115
County 61 92 112 123 388
State 1 0 20 32 53
Multiple 5 3 70 19 97
TOTAL 275 99 234 178 786
* Not all respondents provided demographic data
162
The average respondent has 18 years of experience in the criminal justice profession.
Respondents work in a wide range of jurisdictions including: villages (12.7%), townships (4.2%)
cities (14.6%) counties (49.4%), and state (6.7%); approximately 12 percent of respondents work
in multiple jurisdictions. The ranges of ages of all respondents are reported as: 50-59 years-old
(37.9%), 40-49 years-old (30.0%), 30-39 years-old (16.4%), 60-69 years-old (10.7%), 18-29
years-old (3.2%) and 70-79 years-old (1.9%). The mean-age of respondents is age 49. The
majority of the respondents are male (88%). Caucasians comprise 97.1 percent of the sample,
African Americans 1.4 percent, Hispanic 0.4 percent, Asian 0.1 percent, and 0.8 percent of
respondents identify themselves as “other.”
Research Questions and Statistical Analysis
In this section each research question is stated along with the related survey questions.
The findings are then presented. Frequency distribution charts display bivariate analysis of the
responses of the members of each group to each question. In order to examine group differences,
analysis of variance tests are conducted on all interval-scale items in order to determine if
significant differences (p = .05) exist between group-means. If significant differences are
determined, a post hoc Tukey HSD analysis is completed in order to determine which groupmeans
differ significantly. Only frequencies and means are reported for ordinal-scale data.
Nonparametric chi-square tests which may, under certain circumstances, be conducted on
ordinal-scale data were not used because the data violated chi-square assumptions that (1) no cell
have an observed raw frequency of “zero,” and (2) all expected frequencies but one are least
“five.” (Bartz 1999)
163
Research Question # 1 – How Frequently Does Wrongful Felony Conviction Occur?
To measure the frequency of wrongful conviction, respondents were asked two questions.
First, respondents were asked to estimate the percentage of all cases in the own jurisdiction that
involve wrongful felony conviction. Second, they were asked to estimate the percentage of all
cases in the United States that involve wrongful felony conviction. The second question was
asked because criminal justice professionals often have access to information from outside their
own jurisdiction (meetings and seminars, professional journals, information exchanges, private
conversations and with other professionals, etc.) which should enable them to make educated
estimates concerning events in other jurisdictions.
How frequently do criminal justice professional estimate wrongful conviction occurs in
their own jurisdiction?
To measure perceptions of the extent of wrongful felony conviction in their own
jurisdiction, respondents answered a 10-item Likert scale question with values ranging from
“0%” to “over 25%.” Results are reported in Table 3. Overall, respondents believe
Table 3
Percentages – Wrongful Conviction in Own Jurisdiction
Response Police Prosecutors Def. Attys. Judges All
1 = 0% 33.2 (91)* 29.0 (29) 1.8 (4) 15.5 (26) 19.5 (150)
2 = < than ½% 43.4 (119) 49.0 (49) 2.2 (5) 31.0 (52) 29.3 (225)
3 = ½ % - 1% 13.5 (37) 13.0 (13) 9.3 (21) 21.4 (36) 13.9 (107)
4 = 1% - 3% 6.2 (17) 6.0 (6) 26.5 (60) 19.0 (32) 15.0 (115)
5 = 4% - 5% 3.3 (9) 1.0 (1) 18.6 (42) 6.0 (10) 8.1 (62)
6 = 6% - 10% 0.4 (1) 2.0 (2) 17.3 (39) 5.4 (9) 6.6 (51)
7 = 11% - 15% --- --- 9.3 (21) 1.2 (2) 3.0 (23)
8 = 16% - 20% --- --- 7.1 (16) 0.6 (1) 2.2 (17)
9 = 21% - 25% --- --- 5.3 (12) --- 1.6 (12)
10 = Over 25% --- --- 2.7 (6) --- 0.8 (6)
TOTAL 100.0 (274) 100.0 (100) 100.0 (226) 100.0 (168) 100.0 (768)
MEAN 2.04 2.07 5.33 2.93 3.21
* Number of respondents in parentheses
164
wrongful felony conviction occurs in their own jurisdictions between one-half percent and one
percent of the time. When group responses are compared, defense attorneys perceive higher
rates of wrongful conviction in their jurisdictions than do judges, prosecutors, and police.
Defense attorneys believe that in-jurisdiction wrongful conviction occurs in four to five percent
of all felony cases – several times higher than is estimated by prosecutors, judges, and police
who believe wrongful conviction occurs in their jurisdictions in less that one-half percent of all
felony cases.
Statistically significant differences (p = .000) in the mean scores among the four groups
were determined by analysis of variance. The results of a post hoc Tukey HSD analysis are
reported on Table 4. All group responses significantly differ from one another except those of
prosecutors and police.
Table 4
Wrongful Conviction in Own Jurisdiction
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .998
Defense Attorneys .000 .000
Judges .000 .000 .000
How frequently do criminal justice professional estimate wrongful conviction occurs in the
United States?
To measure beliefs about the extent of wrongful felony conviction in the United States,
respondents answered another 10-item Likert scale question with values ranging from “0%” to
“over 25%.” Table 5 reports the results.
Overall, respondents perceive that wrongful convictions occur in the United States
between one and three percent of the time. This estimate is higher than the estimate given by the
respondents on the previous question concerning their own jurisdictions (½ % - 1%). When
165
group responses are compared, defense attorneys again report higher rates of wrongful
conviction than do judges, prosecutors, and police. Defense attorneys believe the rate of
wrongful conviction in the United States to be between four and five percent (the same as their
in-jurisdiction estimates), while prosecutors and police believe wrongful convictions occur in
one-half to one percent of all felony cases (higher than their in-jurisdiction estimates). Judges
perceive the rate to be slightly greater than prosecutors and police, between one percent and three
percent (also higher than their in-jurisdiction estimates).
Table 5
Percentages – Wrongful Conviction in United States
Response Police Prosecutors Def. Attys. Judges All
1 = 0% 1.1 (3) 1.0 (1) 0.5 (1) --- 0.7 (5)
2 = < than ½% 21.3 (57) 30.2 (29) 1.9 (4) 17.0 (26) 15.8 (116)
3 = ½ % - 1% 23.6 (63) 31.3 (30) 6.0 (13) 20.3 (31) 18.7 (137)
4 = 1% - 3% 29.6 (79) 24.0 (23) 14.4 (31) 28.1 (43) 24.0 (176)
5 = 4% - 5% 14.2 (38) 10.4 (10) 25.9 (56) 18.3 (28) 18.0 (132)
6 = 6% - 10% 5.2 (14) 1.0 (1) 19.4 (42) 10.5 (16) 10.0 (73)
7 = 11% - 15% 2.2 (6) 2.1 (2) 8.8 (19) 3.9 (6) 4.5 (33)
8 = 16% - 20% 2.2 (6) --- 13.0 (28) 1.3 (2) 4.9 (36)
9 = 21% - 25% 0.4 (1) --- 2.3 (5) --- 0.8 (6)
10 = Over 25% --- --- 7.9 (17) 0.7 (1) 2.5 (18)
TOTAL 100.0 (267) 100.0 (96) 100.0 (216) 100.0 (153) 100.0 (732)
MEAN 3.73 3.24 5.91 4.06 4.38
An analysis of variance test indicates statistically significant differences in the mean
scores among the four groups. Analysis of the subgroups again reveals that the responses of
defense attorneys differ significantly (p = .000) from those of the other three groups (Table 6).
In this case, however, judges’ responses differ only from those of prosecutors
(p = .001) and defense attorneys (p = .000). Again, no significant differences are reported in the
police perceptions when compared with those of prosecutors or judges.
166
Table 6 – Frequency of Wrongful Conviction in United States
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .053
Defense Attorneys .000 .000
Judges .174 .001 .000
Research Questions # 2 and # 3. How Frequently Do Criminal Justice Professions
Commit Errors Associated With The Incidence Of Wrongful Conviction, and What is the
General Reliability of the Work Product of Police, Prosecutors, Defense Attorneys, and
Judges?
Police Error
Police error has been found to be associated with cases of wrongful conviction. To
examine the issue of police error, criminal justice professionals were asked three questions. One
question asked respondents for their opinions regarding the frequency of five types of police
error. A second question asked respondents how often they believed police identification
procedures (e.g. showups, photospreads, lineups) contribute to misidentification. A final
question asked respondents to state their perceptions regarding the general reliability of evidence
presented in court by police officials.
How frequently do police officials engage in the following conduct: inadequate
investigation, coaching witnesses in pretrial identification procedures, suppressing
exculpatory evidence, using false evidence, and using undue pressure to obtain a
confession?
167
Each of these five categories is measured using a 9-item Likert scale (see below).
Respondents were asked to indicate, in a range from “never” to “always,” how often they
perceived these types of error to occur. Table 7 reports the mean responses.
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Table 7 – Perceptions of Five Types of Police Error (Mean Responses)
Based on your knowledge and experience,
estimate the frequently of each type of
police error:
Defense All
Type of Police Error Police Prosecutors Attorneys Judges Respondents
Inadequate Police Investigation 3.57 3.93 5.90 4.63 4.55
Police Using Undue Pressure to Obtain a Confession 2.79 2.68 6.23 3.66 4.01
Police Coaching Witnesses in Pretrial I.D. Procedures 2.87 2.48 5.65 3.44 3.78
Police Suppressing Exculpatory Evidence 2.55 2.27 5.52 3.39 3.60
Police Using False Evidence 1.94 1.64 3.71 2.46 2.55
Several patterns are evident when the responses reported in Table 7 are examined. First,
regarding the overall effort of police investigations, except in one instance (Defense Attorneys),
each group of respondents believe that that Inadequate Police Investigation occurs more
frequently that the other types of error. When respondents were directed to consider four types
of error that require police to affirmatively act in an improper manner, Police Using Undue
Pressure to Obtain a Confession is rated highest among prosecutors, defense attorneys, and
judges, and Police Coaching Witnesses in Pretrial I.D. Procedures is rated highest by the police
themselves. All four groups rate Police Using False Evidence as occurring less frequently than
the other three types of affirmative error. All affirmative error is rated by police and prosecutors
to occur “less than infrequently.” Judges rate all affirmative error as occurring “more than
infrequently” (except Police Using False Evidence which is rated by judges to occur “less than
infrequently”) and defense attorneys say all forms of affirmative error occur “more than
moderately frequently” (except Police Using False Evidence which is rated by defense attorneys
to occur “less than infrequently”)
168
Another pattern is evident. Across all five categories, defense attorneys perceive each
type of police error as occurring more frequently than do the other three groups. Mean-response
ratings of defense attorneys are double those of police and prosecutors, and are fifty-percent
higher than those of judges. Prosecutors, on the other hand, in four of the five categories of
police error perceive less error than do police themselves.
A reliability test was conducted to determine if all five items in Table 7 are measuring the
same concept of police error. The Cronbach alpha (.9179) suggests that all five items can be
combined as a single variable. As such, an additive scale was produced and a single variable
“Police Error” created (Table 8).
Table 8
Percentages – Index Variable “Police Error”
Police Prosecutors Def. Attys. Judges All
5=Never
13.9 (188) 21.6 (106) .6 (7) 5.8 (51) 9.0 (352)
10=> Never
<Infrequent
31.9 (432) 31.0 (152) 4.1 (48) 18.2 (159) 20.3 (791)
15=Infrequent
33.5 (454) 28.2 (138) 13.9 (164) 34.2 (299) 27.1 (1055)
20=>Infrequent
<Mod. Freq.
10.9 (148) 10.0 (49) 10.5 (124) 15.9 (139) 11.8 (460)
25=Moderately
Frequent
7.5 (101) 5.5 (27) 25.7 (303) 17.1 (149) 14.9 (580)
30=>Mod Freq
<Very Freq
1.5 (20) 1.6 (8) 10.5 (124) 3.9 (34) 4.8 (186)
35=Very
Frequent
0.9 (12) 1.6 (9) 23.7 (279) 4.2 (37) 8.7 (337)
40=>Very Freq
<Always
--- .2 (1) 8.4 (99) .6 (5) 2.7 (105)
45=Always
--- --- 2.5 (29) --- 0.7 (29)
TOTAL 100.0
(1355)
100.0
(490)
100.0
(1177)
100.0
(873)
100.0
(3895)
MEAN 13.17 13.00 27.00 17.58 18.50
169
Mean-response to the variable Police Error (18.50) indicates respondents believe this type
of error to occur slightly above the “infrequent” level. Fifty-seven percent of all respondents
believe Police Error occurs “infrequent,” or less, while forty-three percent of all respondents
perceive police error to occur more than “infrequent.” When group responses are analyzed
separately, defense attorneys think Police Error occurs more often than do prosecutors and police
(moderately frequent vs. less than infrequent). Judges perceive police error to occur more
frequently than do prosecutors and police, but less frequently than do defense attorneys.
An analysis of variance test of the variable Police Error indicates that statistically
significant differences (p = .000) in the mean-scores of the four groups. The results of the post
hoc Tukey HSD test are shown in Table 9. Responses of defense attorneys are significantly
different (p = .000) than members of the other three groups. Judges’ responses also significantly
differ (p = .000) from those of the other three groups. The responses of police and prosecutors
do not differ significantly.
Table 9
Additive Scale Variable “Police Error”
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .808
Defense Attorneys .000 .000
Judges .000 .000 .000
How often do police identification procedures contribute to misidentification?
Perceived frequency of error contributed to by police identification procedures is
measured using a 4-item Likert scale. Table 10 reports the findings. Overall, respondents
believe that this type of error occurs slightly more than “seldom.” When examining the
responses by group, the great majority of police, prosecutors and judges perceive that police
identification procedures “seldom” contribute to misidentification. However, only about one170
third of defense attorneys selected the “seldom” response. Instead, mean response of defense
attorneys were closer to “often,” with over half (54.9%) of the defense attorneys selecting the
“often” response – which is several times higher than prosecutors, police and judges.
Table 10 –Percent - Police Identification Procedures Contribute to Misidentification
Response Police Prosecutors Def. Attys Judges All
1 = Never 4.1 (11) 8.0 (8) 1.3 (3) 4.6 (8) 3.9 (30)
2 = Seldom 92.3 (250) 89.0 (89) 33.0 (77) 84.0 (147) 72.3 (563)
3 = Often 3.7 (10) 2.0 (2) 54.9 (128) 11.4 (20) 20.5 (160)
4 = Very Often --- 1.0 (1) 10.7 (25) --- 3.3 (26)
TOTAL 100.0 (271) 100.0 (100) 100.0 (233) 100.0 (175) 100.0 (779)
MEAN 2.00 1.96 2.75 2.07 2.23
In general, how reliable is the evidence presented in court by police officials?
The reliability of evidence presented in court by police is measured using a 4-item Likert scale
asking respondents to indicate whether they think such evidence is “very reliable,” “usually
reliable,” “usually unreliable,” or “very unreliable.” Response frequencies and group means are
reported in Table 11. General reliability of police evidence is perceived by all respondents to be
slightly more than “usually reliable.” When group differences are examined, the largest
differences in responses occur in the “very reliable” category. While the majority of police and
prosecutors say police evidence is “very reliable,” only 33.4 percent of judges and 1.8 percent of
defense attorneys select this response. Also, approximately one third of defense attorneys select
the “usually unreliable” response, as compared to only 1.0 percent of prosecutors, 1.8 percent of
judges, and 1.9 percent of prosecutors. In sum, while the majority of all respondents perceive the
general reliability of police evidence to be “usually reliable,” police officials viewed their own
groups’ evidence as more reliable than prosecutors, judges, and defense attorneys.
171
Table 11
Percentages – General Reliability of Police Evidence
Response Police Prosecutors Def. Attys Judges All
1 = Very
Reliable
62.0 (160) 55.6 (55) 1.8 (4) 18.3 (31) 33.4 (250)
2 = Usually
Reliable
35.3 (91) 43.4 (43) 60.5 (135) 79.3 (134) 53.8 (403)
3 = Usually
Unreliable
1.9 (5) 1.0 (1) 35.0 (78) 1.8 (3) 11.6 (87)
4 = Very
Unreliable
0.8 (2) --- 2.7 (6) 0.6 (1) 1.2 (9)
TOTAL 100.0 (258) 100.0 (99) 100.0 (223) 100.0 (169) 100.0 (749)
MEAN 1.41 1.45 2.39 1.85 1.81
Prosecutorial Error
Previous research suggests that prosecutorial error is associated with cases of wrongful
conviction. To examine this issue, respondents were asked two questions. One question, asked
respondents their opinions regarding to indicate their perceptions regarding the frequency of five
types of prosecutorial error. Respondents were also presented with a question asking them their
opinions regarding the general reliability of evidence presented in court by prosecutors.
How frequently do prosecutors engage in the following conduct: inadequate
investigation of case, suppressing exculpatory evidence, using undue plea bargaining
pressure, prompting a witness, knowingly using false testimony?
These five categories are measured using a 9-item Likert scale (see below).
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Respondents were asked to indicate, in a range from “never” to “always,” how often they
believed each type of error occurs. Mean responses of all four groups of criminal justice
professionals, individually and compositely, are reported in Table 12.
172
Table 12
Perceptions of Five Types of Prosecutorial Error (Mean Responses)
Based on your knowledge and experience,
estimate the frequently of each type of
prosecutorial error:
Defense All
Type of Prosecutorial Error Police Prosecutors Attorneys Judges Respondents
Inadequate Investigation of Case by Prosecutor 3.67 2.67 5.28 4.03 4.11
Prosecutor Using Undue Plea Bargain Pressure 3.94 2.20 5.65 3.57 4.15
Prosecutor Prompting a Witnesses 3.27 2.22 5.83 3.73 4.01
Prosecutor Suppressing Exculpatory Evidence 2.67 1.59 4.66 2.88 3.18
Prosecutor Knowingly Using False Testimony 1.84 1.25 3.11 1.96 2.17
An examination of Table 12 reveals that each group of respondents agree that Prosecutor
Knowingly Using False Evidence occurs the least frequently of the five types of error.
Inadequate Prosecutorial Investigation was indicated by prosecutors and judges to occur the most
frequently of the five types of error. (However, prosecutors indicated Inadequate Investigation
occurs “less than infrequent,” while judges indicated that it occurs “between “infrequent: and
“moderately frequent.”) Police thought that Prosecutors Using Undue Plea Bargaining Pressure
occurred the most often of the five types of error (between “infrequent” and “moderately
frequent”), and defense attorneys perceived Prosecutors Prompting Witnesses as occurring the
most often (between “moderately frequent” and “very frequent”). Defense attorneys perceived
all types of prosecutorial error as occurring more often than any of the other three groups.
Prosecutors, on the other hand, perceived that their own group commits less error than is
perceived by the other three groups.
A reliability test was conducted to determine if all five items in Table 12 are measuring
the same concept of prosecutor error. A high Cronbach alpha (.8998) suggests that all five items
can be included into a single variable. As such, an additive scale was produced (Table 13) using
all five survey items, and a single variable “Prosecutor Error” was created.
173
Table 13
Percentages – Index Variable “Prosecutor Error”
Police Prosecutors Def. Attys. Judges All
5=Never
12.6 (171) 39.9 (199) 2.0 (24) 12.3 (107) 12.8 (501)
10=> Never
<Infrequent
26.6 (359) 31.7 (158) 8.2 (97) 22.3 (194) 20.7 (808)
15=Infrequent
31.1 (420) 21.0 (105) 19.5 (230) 32.5 (283) 26.6 (1038)
20=>Infrequent
<Mod. Freq.
11.8 (159) 5.2 (26) 11.4 (134) 12.3 (107) 10.9 (426)
25=Moderately
Frequent
10.9 (148) 1.8 (9) 22.5 (265) 13.0 (113) 13.7 (535)
30=>Mod Freq
<Very Freq
2.3 (35) .4 (2) 9.2 (109) 2.8 (24) 4.4 (170)
35=Very
Frequent
4.0 (54) --- 19.4 (229) 4.0 (35) 8.3 (318)
40=>Very Freq
<Always
.4 (5) --- 5.2 (61) .7 (6) 1.8 (72)
45=Always
.1 (1) --- 2.5 (30) .1 (1) 0.8 (32)
TOTAL 100.0
(1352)
100.0
(499)
100.0
(1179)
100.0
(870)
100.0
(3900)
MEAN 15.39 9.93 24.53 15.99 17.59
The mean-response for all respondents for the variable Prosecutor Error is slightly
above the midpoint between “infrequent” and “more than infrequent but less than moderately
frequent” (mean = 17.59). Sixty percent of all respondents are of the opinion that Prosecutor
Error occurs at the “infrequent” or less level, and forty percent of respondents believe that
Prosecutor Error occurs “more than infrequently.” When the groups are analyzed separately, the
major differences in perceptions are best understood when the responses are broken down into
“infrequent or less” and “more than infrequent” categories. Only 29.7 percent of defense
attorneys perceive Prosecutor Error to occur at the “infrequent or less” level. On the other hand,
prosecutors selected responses of “infrequent or less,” at a rate three times higher than do
defense attorneys, and two times higher than did judges and police. Conversely, 60.3 percent of
174
defense attorneys perceive Prosecutor Error to occur “more than infrequent” – almost double the
perceptions for the same categories as those of judges and police, and is seven times higher than
perceived by prosecutors.
An analysis of variance test of the variable Prosecutor Error reveals statistically
significant (p = .000) differences in mean scores of the four groups. The results of a post hoc
Tukey HSD test are shown in Table 14. Defense attorneys perceive a significantly (p = .000)
different rate of Prosecutor Error than do members of the other three groups. Prosecutor
responses also differ from the other three groups (p =.000), while judges and police responses
perceptions do not differ.
Table 14
Additive Scale Variable “Prosecutor Error”
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .000
Defense Attorneys .000 .000
Judges .315 .000 .000
In general, how reliable do criminal justice professionals believe is the evidence
presented in court by prosecutors?
Perceived reliability of evidence presented in court by prosecutors is measured using a 4-
item Likert scale. Table 15 reports the findings. When all responses are
examined respondents perceive the general reliability of prosecutorial evidence to be slightly
above “usually reliable.” When group differences are examined, the largest differences occur in
the “very reliable” category. While a large majority of prosecutors (74.5%) perceive their own
evidence to be “very reliable,” as do a near majority of police (45.9%), only about one-fourth of
judges (25.9%) perceive prosecutorial evidence to be “very reliable.” Most judges perceive
175
prosecutorial evidence to be “usually reliable.” Defense attorneys have dramatically different
views from those of the other three groups. Only 1.3 percent of defense attorneys indicate that
prosecutorial evidence is “very reliable,” and 20.7 percent believe the evidence to be “usually
unreliable.”
Table 15
Percentages – General Reliability of Prosecutorial Evidence
Response Police Prosecutors Def. Attys Judges All
1 = Very
Reliable
45.9 (123) 74.5 (73) 1.3 (3) 25.9 (44) 31.6 (243)
2 = Usually
Reliable
51.9 (139) 25.5 (25) 78.0 (181) 72.9 (124) 61.1 (469)
3 = Usually
Unreliable
1.1 (3) --- 20.7 (48) 0.6 (1) 6.8 (52)
4 = Very
Unreliable
1.1 (3) --- --- 0.6 (1) 0.5 (4)
TOTAL 100.0 (268) 100.0 (98) 100.0 (232) 100.0 (170) 100.0 (768)
MEAN 1.57 1.26 2.19 1.76 1.76
Defense Attorney Error
Previous research indicates that defense attorney error is associated with cases of
wrongful conviction. To examine this issue, criminal justice professionals were asked two
questions. One question asked respondents to indicate their perceptions regarding the frequency
of five types of defense attorney error. Respondents are also presented with a question asking
them their opinions regarding how well they believe defense attorneys defend their clients.
How frequently do defense attorneys engage in the following conduct: inadequate
investigation of case, failing to file proper motions, not adequately challenging forensic
evidence, not adequately challenging witnesses, making unwarranted plea bargain
concessions?
These five categories are measured using a 9-item Likert scale (see below) asking
respondents to indicate, in a range from “never” to “always,” how often the perceive these
176
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
types of error to occur. Mean response of all four groups of criminal justice professionals,
individually and compositely, are reported in Table 16. Of the five categories of defense
attorney error, all respondents perceive that Inadequate Investigation of Case by Defense
Attorney occurs the most frequently. Of the four types of affirmative error, police believe that
Defense Attorney Making Unwarranted Plea Bargain Concessions occurs the most frequently,
and prosecutors think Defense Attorney Failing to File Proper Motions occurs most often.
Judges, and the defense attorneys themselves, believe that Not Adequately Challenging Forensic
Evidence occur with the most frequency. The least frequently occurring error, according to
prosecutors, judges, and defense attorneys is Defense Attorney Making Unwarranted Plea
Bargain Concessions, while police believe that Defense Attorneys not Adequately Challenging
Witnesses occurs the least often. Defense attorneys rate their own error, in all five categories,
higher than do members of the other groups. Of interest is the fact that prosecutors rate defense
attorney error lower than is rated by members of the other three groups.
Table 16
Perceptions of Five Types of Defense Error (Mean Responses)
Based on your knowledge and experience,
estimate the frequently of each type of
defense attorney error:
Defense All
Type of Defense Attorney Error Police Prosecutors Attorneys Judges Respondents
Inadequate Investigation of Case by Defense Attorney 4.54 4.31 5.35 4.70 4.79
Defense Attorney Not Adequately Challenging Forensic Evidence 3.84 3.71 5.24 4.07 4.30
Defense Attorney Failing to File Proper Motions 4.02 3.87 4.69 4.02 4.21
Defense Attorney Making Unwarranted Plea Bargain Concessions 4.30 3.12 4.59 3.85 4.14
Defense Attorney Not Adequately Challenging Witnesses 3.68 3.43 4.61 3.96 3.99
A reliability test was conducted to determine if all five items in Table 16 were measuring
the same concept of “defense attorney error.” A high Cronbach alpha (.8879) suggests that all
five variables can be included in a single variable. As such, an additive scale was produced
177
using the five survey items, and a single variable “Defense Attorney Error” was created (Table
17).
Table 17
Percentages – Index Variable “Defense Attorney Error”
Police Prosecutors Def. Attys. Judges All
5=Never
.8 (11) 4.6 (23) .3 (4) .7 (6) 1.1 (44)
10=> Never
<Infrequent
9.5 (129) 15.0 (75) 4.0 (48) 9.2 (82) 8.5 (334)
15=Infrequent
32.3 (440) 37.0 (185) 17.8 (211) 31.1 (278) 28.3 (1114)
20=>Infrequent
<Mod. Freq.
22.1 (301) 14.6 (73) 16.1 (191) 18.9 (168) 18.6 (733)
25=Moderately
Frequent
20.1 (273) 16.4 (82) 30.8 (366) 26.1 (233) 24.2 (954)
30=>Mod Freq
<Very Freq
6.3 (86) 7.0 (35) 11.3 (134) 5.9 (53) 7.8 (308)
35=Very
Frequent
7.1 (97) 4.2 (21) 16.1 (191) 7.5 (67) 9.5 (376)
40=>Very Freq
<Always
1.5 (20) 1.0 (5) 3.3 (39) .8 (7) 1.8 (71)
45=Always
.3 (4) .2 (1) .3 (4) --- 0.2 (9)
TOTAL 100.0
(1361)
100.0
(500)
100.0
(1188)
100.0
(894)
100.0
(3943)
MEAN 20.39 18.36 24.51 20.60 21.44
Overall, the mean-response for Defense Attorney Error is below the “moderately
frequent” level (mean = 21.44). There is a large consensus among prosecutors, police, and
judges that Defense Attorney Error occurs below “moderately frequently.” Only defense
attorneys believe their group’s own error to be at the “moderately frequent” level.
An analysis of variance test reveals that significant differences (p = <.05) exist in the
mean scores of the four groups. Table 18 reports the results of a post hoc Tukey HSD test.
Defense attorneys perceive significantly (p =.000) different rates of their own group’s error than
do members of the other three groups. Prosecutors report significantly different rates of Defense
178
Attorney Error than do police (p = .039) and judges (p = .025). Police and judges’ perceptions
do not significantly differ.
Table 18
Additive Scale Variable “Defense Attorney Error”
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .039
Defense Attorneys .000 .000
Judges .973 .025 .000
In general, how well do criminal justice professionals believe that defense attorneys
defend their clients?
Table 19 reports the findings of a 4-item Likert scale question which asked respondents
to indicate how well they believe defense attorneys defend their clients.
While respondents from all groups believe defense attorneys defend their clients “moderately
well,” defense attorneys are more critical of their own group than are prosecutors, judges, and
police. The differences in group perceptions are most evident in the “very well” category.
Approximately one in ten defense attorneys rate their own groups’ defense of clients as “very
well,” compared to about one in four prosecutors and judges. Across all groups, however, a
majority of respondents agree that defense attorneys defended their clients “moderately well.”
Table 19
Percentages – How Well Defense Attorneys Defend Clients
Response Police Prosecutors Def. Attys Judges All
1 = Very
Well
15.5 (41) 28.3 (28) 11.1 (26) 25.6 (43) 18.0 (138)
2 = Moderately
Well
72.3 (191) 64.6 (64) 76.6 (180) 67.9 (114) 71.7 (549)
3 = Not Very
Well
11.0 (29) 7.1 (7) 11.9 (28) 6.5 (11) 9.8 (75)
4 = Poorly
1.1 (3) --- 0.4 (1) --- 0.5 (4)
TOTAL 100.0 (264) 100.0 (99) 100.0 (235) 100.0 (168) 100.0 (766)
179
MEAN 1.98 1.79 2.02 1.81 1.93
Judicial Error
Previous research contends that judicial error is associated with cases of wrongful
conviction. To examine this issue, criminal justice professionals were asked two questions. One
question asked respondents to indicate their perceptions regarding the frequency of four types of
judicial error. Respondents were also presented with a question asking them their opinions
regarding how reliable they believe are judicial decisions regarding defendants’ guilt or
innocence.
How frequently do judges engage in the following conduct: error concerning
admissibility of physical evidence, error concerning admissibility of eyewitness testimony,
error concerning admissibility of expert testimony, error resulting from judicial bias?
These four categories are measured using a 9-item Likert scale (see below) asking
respondents to indicate, in a range from “never” to “always,” how often they perceive these types
of error to occur.
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Mean response rates of all four groups of criminal justice professionals, individually and
compositely, are reported in Table 20.
Table 20
Perceptions of Four Types of Judicial Error (Mean Responses)
Based on your knowledge and experience,
estimate the frequently of each type of
judicial error:
Defense All
Type of Judicial Error Police Prosecutors Attorneys Judges Respondents
Error Concerning Admissibility of Physical Evidence 3.00 2.87 4.55 3.00 3.45
Error Resulting From Judicial Bias 2.90 2.36 5.07 2.69 3.44
Error Concerning Admissibility of Eyewitness Testimony 3.01 2.50 4.45 2.84 3.35
Error Concerning Admissibility of Expert Testimony 2.79 2.54 4.25 2.94 3.23
180
Overall, the respondents report that “error concerning admissibility of physical evidence”
occurs the most often – slightly above the “infrequent” level. When comparing across the four
groups, mean responses of prosecutors are below the “infrequent” level, and mean responses of
police and judges are at the “infrequent” level. Defense attorneys say this type of judicial error
occurs near the “moderately frequent” level.
The second category, “error resulting from judicial bias,” is also perceived to occur more
than infrequently. When comparing across the four groups, mean responses of prosecutors,
police, and the judges themselves are below the “infrequent” level, while defense attorneys are of
the opinion that judicial bias occurs above the “moderately frequent” level – with a meanresponse
more than double prosecutors and approximately 40 percent higher than judges.
Perceptions of “error concerning admissibility of eyewitness testimony,” follow along
similar lines. Overall, respondents say this type of error occurs above the “infrequent” level.
Comparing the four groups, prosecutors feel this type of error occurs less than “infrequent,”
while police and judges say error concerning admissibility of eyewitness testimony occurs more
than “infrequent.” Again, defense attorneys believe this type of error occurs at a higher level
than do the other three groups, at slightly below the “moderately frequent” level.
The final category, “error concerning admissibility of expert testimony,” is viewed by all
respondents as occurring more than infrequently. When responses of the four groups are
compared, the pattern slightly changes. Prosecutors, police, and judges all say this type of error
occurs below the “infrequent” level, while defense attorneys believe this occurs at below the
“moderately frequent” level.
For every category there appears to be a basic agreement among prosecutors, police, and
judges that judicial error is “infrequent” or less. Defense attorneys, on the other hand, perceive
181
judicial error as occurring either at or below the “moderately frequent” level. Of interest is the
fact that prosecutors, in each of the four categories, estimate less
error for judges than do judges themselves. Police also report less error for judges than do the
judges themselves in the category “error concerning admissibility of expert testimony.”
A reliability test was conducted to determine if all four items in Table 20 are measuring
the same concepts of “judicial error.” A high Cronbach alpha (.8772) suggests that all four items
can be combined as a single variable. As such, an additive scale was produced using the four
survey items and a new variable “Judicial Error” was created (Table 21).
Overall, the mean-response for the variable Judicial Error is above the “infrequent” level
(mean = 13.41). There is basic consensus among prosecutors, police, and judges that judicial
error occurs at the less than “infrequent” level. Defense attorneys perceive Judicial Error to
occur slightly below the “moderately frequent” level.
Table 21
Percentages – Index Variable “Judicial Error”
Police Prosecutors Def. Attys. Judges All
4=Never
7.0 (76) 10.8 (43) .1 (1) 3.7 (25) 4.7 (145)
8=> Never
<Infrequent
33.4 (363) 40.0 (160) 6.0 (56) 30.7 (210) 25.5 (789)
12=Infrequent
37.4 (407) 36.5 (146) 28.0 (260) 46.6 (318) 36.5 (1131)
16=>Infrequent
<Mod. Freq.
12.0 (130) 8.3 (33) 18.0 (167) 9.8 (67) 12.8 (397)
20=Moderately
Frequent
7.3 (79) 3.5 (14) 23.4 (217) 6.3 (43) 11.4 (353)
24=>Mod Freq
<Very Freq
1.3 (14) 1.0 (4) 10.1 (94) 1.2 (8) 3.9 (120)
28=Very
Frequent
1.5 (16) --- 9.6 (89) 1.8 (12) 3.8 (117)
32=>Very Freq
<Always
.2 (2) --- 3.9 (36) --- 1.2 (38)
36=Always
--- --- .9 (8) --- .9 (8)
TOTAL 100.0 100.0 100.0 100.0 100.0
182
(1087) (400) (928) (683) (3098)
MEAN 11.59 10.27 18.07 11.80 13.41
An analysis of variance test of the variable Judicial Error reveals that significant
differences exist in the group mean scores. Table 22 reports the results of a post hoc Tukey HSD
test. Defense attorneys perceive a significantly (p=.000) different rate of judicial error than do
members of the other three groups. Prosecutors, police, and judges’ perceptions do not differ
significantly.
Table 22
Additive Scale Variable “Judicial Error”
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .110
Defense Attorneys .000 .000
Judges .937 .348 .000
In general, how reliable do criminal justice professionals believe are judicial decisions
concerning defendants’ guilt or innocence?
Perceived reliability of judicial decisions concerning guilt or innocence is measured using
a 4-item Likert scale. Table 23 reports the findings. The mean-response of all groups indicates
perceptions regarding judicial decisions concerning guilt or innocence to be above the “usually
reliable” level. When individual group responses are investigated, prosecutors, judges, and
police all believe these types of judicial decisions are between “very reliable” and “usually
reliable” (although closer to “usually reliable” than “very reliable”). Defense attorneys are more
critical of judges, and find the quality of judicial decisions concerning guilt or innocence to be
between “usually reliable” and “moderately reliable.”
183
Table 23
Percentages – Reliability of Judicial Decisions
Concerning Defendants’ Guilt or Innocence
Response Police Prosecutors Def. Attys Judges All
1 = Very
Reliable
39.6 (107) 53.0 (53) 3.0 (7) 45.5 (81) 31.7 (248)
2 = Usually
Reliable
50.4 (136) 42.0 (42) 47.7 (112) 48.3 (86) 48.0 (376)
3 = Moderately
Reliable
9.6 (26) 5.0 (5) 43.8 (103) 6.2 (11) 18.5 (145)
4 = Unreliable
0.4 (1) --- 5.5 (13) --- 1.8 (14)
TOTAL 100.0 (270) 100.0 (100) 100.0 (235) 100.0 (178) 100.0 (783)
MEAN 1.71 1.52 2.52 1.61 1.90
Research Questions # 4 and #5. How Frequently Do Non-Criminal Justice Personnel
Commit Errors Associated with the Incidence of Wrongful Conviction and What is the
General Reliability of Eyewitness and Forensic Testimony?
Eyewitness Error
Previous research suggests that eyewitness error is associated with cases of wrongful
conviction. To examine this issue, respondents are asked to indicate their perceptions
concerning three survey questions.
How frequently do eyewitnesses, in good faith, misidentify a defendant?
Perceived frequency of good-faith eyewitness error is measured using a 4-item Likert
scale question. Respondents were asked to indicate whether they believe this type of error
occurs “never,” “seldom,” “often,” or “very often.” Error is perceived by all
respondents to occur slightly above the “seldom” range (Table 24). A substantial majority
(79.8% to 89.2%) of police, prosecutors, and judges say good-faith eyewitness error
184
Table 24
Percentages – Good-faith Eyewitness Misidentification
Response Police Prosecutors Def. Attys Judges All
1 = Never 0.7 (2) 4.9 (5) 0.4 (1) 4.6 (8) 2.0 (16)
2 = Seldom 85.0 (232) 89.2 (91) 41.2 (98) 79.8 (138) 71.2 (559)
3 = Often 14.3 (39) 4.9 (5) 50.8 (121) 15.0 (26) 24.3 (191)
4 = Very Often --- 1.0 (1) 7.6 (18) 0.6 (1) 2.5 (20)
TOTAL 100.0 (273) 100.0 (102) 100.0 (238) 100.0 (173) 100.0 (786)
MEAN 2.14 2.02 2.66 2.12 2.27
“seldom” occurs. In contrast, defense attorneys are essentially split in their opinions –
approximately 41 percent of defense attorneys perceive good-faith eyewitness testimony to
“seldom” occur, while approximately 59% think it occurs “often” or “very often.”
How frequently do eyewitnesses intentionally misidentify a defendant?
Frequency of intentional eyewitness misidentification is measured using another 4-item
Likert scale. This type of eyewitness error is perceived by all respondents to occur at
Table 25
Percentages – Intentional Eyewitness Misidentification
Response Police Prosecutors Def. Attys Judges All
1 = Never 8.4 (23) 25.5 (26) 3.8 (9) 19.4 (34) 11.7 (92)
2 = Seldom 88.3 (242) 73.5 (75) 79.4 (189) 80.0 (140) 81.9 (646)
3 = Often 3.3 (9) 1.0 (1) 16.0 (38) 0.6 (1) 6.2 (49)
4 = Very Often --- --- 0.8 (2) --- 0.3 (2)
TOTAL 100.0 (276) 100.0 (102) 100.0 (240) 100.0 (180) 100.0 (789)
MEAN 1.95 1.75 2.14 1.81 1.95
slightly below the “seldom” range (Table 25). Defense attorneys believe intentional eyewitness
misidentification occurs more often than prosecutors, law enforcement, and judges, but still only
slightly above the “seldom” level. Only 6.5 percent of all respondents report that intentional
eyewitness misidentification occurs “often” or “very often” – primarily due to the opinions of
185
defense attorneys. Approximately one-fourth of prosecutors and one-fifth of judges say that
intentional misidentification “never” occurs.
In general, how reliable is eyewitness testimony?
Table 26 reports respondents’ perceptions of the reliability of eyewitness testimony.
Table 26
Percentages – General Reliability of Eyewitness Testimony
Response Police Prosecutors Def. Attys Judges All
1 = Very
Reliable
18.6 (51) 35.6 (36) 1.7 (4) 16.0 (28) 15.1 (119)
2 = Usually
Reliable
73.0 (200) 59.4 (60) 57.6 (136)
78.9 (138)
67.9 (534)
3 = Usually
Unreliable
7.7 (21)
3.0 (3) 37.7 (89) 4.6 (8) 15.4 (121)
4 = Very
Unreliable
.7 (2) 2.0 (2) 3.0 (7)
.6 (1) 1.5 (12)
TOTAL 100.0 (274) 100.0 (101) 100.0 (236) 100.0 (175) 100.0 (786)
MEAN 1.91 1.71 2.42 1.90 2.03
Overall, a great majority of the respondents report the reliability of eyewitness testimony to be
“usually reliable” (67.9%). When group differences are examined, defense attorneys find
eyewitness testimony to be significantly (p=.000) less reliable then is perceived by prosecutors,
judges, and police. Approximately forty percent of defense attorneys perceive the general
reliability of eyewitness testimony to be “usually unreliable” or “very unreliable” – a rate twelve
times higher than prosecutors, five times higher than police, and twice as high as judges.
Forensic Error
Forensic error has also been found to be associated with cases of wrongful conviction.
To examine the issue of forensic error, respondents are asked to indicate their perceptions
concerning three survey questions.
How frequently do forensic experts make good-faith error?
186
Good-faith error by forensic experts is measured using a 9-item Likert scale (see below)
asking respondents to indicate, in a range from “never” to “always,” how often they believe
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
this type of error occurs. Response frequencies are reported on Table 27. Overall, respondents
perceive this type of error to occur at slightly below the “infrequent” level. Approximately onefourth
of all respondents, however, say this type of error occurs more than “infrequent.” Defense
attorneys report that good-faith forensic error occurs more often than do prosecutors, police, and
judges. A majority of defense attorneys indicate that this type of error occurs more than
“infrequent,” and almost one-fourth (24.7%) of defense attorneys report that good-faith forensic
error occurs “moderately frequent” or more.
Table 27
Percentages – Good-Faith Errors by Forensic Experts
Police Prosecutors Def. Attys. Judges All
1=Never
5.2 (14) 15.7 (16) .4 (1) 5.6 (10) 5.2 (41)
2=> Never
<Infrequent
42.4 (115)
41.2 (42)
6.5 (15)
26.0 (46)
27.9 (218)
3=Infrequent
41.0 (111) 41.2 (42) 41.6 (96) 54.8 (97) 44.3 (346)
4=>Infrequent
<Mod. Freq.
9.2 (25)
1.0 (1)
26.8 (62)
9.6 (17)
13.4 (105)
5= Moderately
Frequent
.7 (2)
1.0 (1)
20.8 (48)
3.4 (6)
7.3 (57)
6=>Mod Freq
<Very Freq
.4 (1)
---
2.2 (5)
---
.8 (6)
7=Very
Frequent
1.1 (3)
---
1.3 (3)
---
.8 (6)
8=>Very Freq
<Always
--- ---
.4 (1)
---
.1 (1)
9=Always
---
---
---
.6 (1) .1 (1)
TOTAL 100.0 (271) 100.0 (102) 100.0 (231) 100.0 (177) 100.0 (781)
MEAN 2.63 2.30 3.75 2.82 2.96
187
Statistically significant differences (p = <.05) in the mean scores among the four groups
were determined by ANOVA. The results of a post hoc Tukey HSD analysis are reported in
Table 28. Defense attorneys report significantly (p = .000) different levels of good-faith
Table 28
Good-Faith Error by Forensic Experts
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .015
Defense Attorneys .000 .000
Judges .165 .000 .000
forensic error than do the other three groups. Likewise, the perceptions of prosecutors and
judges differ significantly (p = .000), as do the views of prosecutors and police (p = .015). The
views of judges and police do not differ significantly.
How frequently do forensic experts intentionally misrepresent evidence?
Perceptions of intentional misrepresentation by forensic experts is also measured using
the same 9-item Likert scale asking respondents to indicate how often they believe this type of
error occurs. Table 29 presents the findings. Intentional forensic error is perceived to occur by
all respondents at the more than “never” but less than “infrequent” level. Defense attorneys
believe this error occurs more often than do prosecutors, police, and judges, though most all
respondents say that intentional misrepresentation occurs infrequently – or less than infrequently.
Approximately one-half of prosecutors (55.9%) believe this type of error “never” occurs – as do
approximately one-third of police (32%) and judges (31.1%). In contrast, only 7 percent of
defense attorneys say that intentional forensic misrepresentation “never” occurs.
188
Table 29
Percentages – Intentional Misrepresentation of Evidence by Forensic Experts
Police Prosecutors Def. Attys. Judges All
1=Never
32.0 (86) 55.9 (57) 7.0 (16) 31.1 (55)
27.5 (214)
2=> Never
<Infrequent
42.8 (115)
35.3 (36)
33.9 (78)
44.1 (78)
39.5 (307)
3=Infrequent
22.7 (61)
8.8 (9)
38.3 (88)
21.5 (38)
35.2 (196)
4=>Infrequent
<Mod. Freq.
1.1 (3)
---
8.7 (20)
1.1 (2)
3.2 (25)
5= Moderately
Frequent
.7 (2)
---
10.0 (23)
2.3 (4)
3.7 (29)
6=>Mod Freq
<Very Freq
.7 (2)
---
.9 (2)
---
.5 (4)
7=Very
Frequent
---
---
.9 (2)
---
.3 (2)
8=>Very Freq
<Always
---
---
---
---
---
9=Always
--- ---
.4 (1) ---
.1 (1)
TOTAL 100.0 (269) 100.0 (102) 100.0 (230) 100.0 (177) 100.0 (778)
MEAN 1.98 1.53 2.90 1.99 2.20
An analysis of variance test reveals that there are significant statistical differences in
responses across the four groups (Table 30). When perceptions of the four groups of respondents
are compared defense attorneys report significantly (p = .000) different mean-responses than do
Table 30
Intentional Misrepresentation of Evidence by Forensic Experts
Significance Levels (Tukey HSD) for Comparison of Group Responses
Police Prosecutors Defense Attys
Prosecutors .000
Defense Attorneys .000 .000
Judges .999 .001 .000
the other three groups. Also, prosecutors’ views differ significantly from those of judges (p =
.001) and police (p = .000). Judges and police shared the view that intentional misrepresentation
189
of evidence by forensic experts occurred more frequently than perceived by prosecutors, but less
often than thought by defense attorneys.
In general, how reliable are the scientific conclusions of forensic experts?
Table 31 reports respondents’ beliefs concerning the general reliability of the conclusions
of forensic experts. Almost all respondents (95.1%) stated that the conclusions of forensic
Table 31
Percentages – Reliability of Scientific Conclusions of Forensic Experts
Response Police Prosecutors Def. Attys Judges All
1 = Very
Reliable
61.8 (168)
74.5 (76)
8.2 (19)
35.2 (62)
41.6 (325)
2 = Usually
Reliable
33.8 (92) 24.5 (25) 81.5 (189)
63.6 (112)
53.5 (418)
3 = Usually
Unreliable
2.2 (6)
--- 10.3 (24) --- 3.8 (30)
4 = Very
Unreliable
2.2 (6) 1.0 (1) ---
1.1 (2) 1.2 (9)
TOTAL 100.0 (272) 100.0 (102) 100.0 (232) 100.0 (176) 100.0 (782)
MEAN 1.45 1.27 2.02 1.67 1.65
experts are “usually reliable” or “very reliable.” Although most respondents said these
conclusions are reliable, defense attorneys and judges perceived that the conclusions are “usually
reliable,” while prosecutors and police said the conclusions are “very reliable.” When looking
only at the opinions of defense attorneys, they perceive these types of conclusions as being less
reliable than is believed by members of the three other groups. Only one in twelve defense
attorneys report the general reliability of scientific conclusions by forensic experts to be “very
reliable.”
190
False Accusations
To examine this issue the following question was presented:
How frequently do false accusations occur?
Perceived frequency of false accusations is measured using a 3-item Likert scale asking
respondents to indicate whether they believe this type of error occurs:
(a) seldom, (b) occasionally, or (c) frequently. Frequencies, response percentages, and means are
reported in Table 32. The frequency of false accusations is viewed by all respondents to occur at
Table 32
Percentages – False Accusation
Response Police Prosecutors Def. Attys. Judges All
1=Seldom 38.8 (106) 53.0 (53) 5.1 (12) 31.8 (56) 28.9 (227)
2= Occasionally 53.5 (146) 44.0 (44) 54.2 (128) 59.7 (105) 53.9 (423)
3=Frequently 7.7 (21) 3.0 (3) 40.7 (96) 8.5 (15) 17.2 (135)
TOTAL 100.0 (273) 100.0 (100) 100.0 (236) 100.0 (176) 100.0 (785)
MEAN 1.69 1.50 2.36 1.77 1.88
slightly below the “occasional” level. When group differences are analyzed, prosecutors are of
the opinion that false accusation occurs the least often – at the midpoint between “seldom” and
“occasionally,” while police and judges believe false accusations occur slightly more often.
Defense attorneys think false accusations occur between “occasionally” and “frequently.”
Differences between the responses of defense attorneys and the other three groups are
highlighted in the “frequently” category. Over 40 percent (40.7%) of defense attorneys perceive
false accusations occur “frequently,” in comparison to only 3 percent of prosecutors, 7.7 percent
of police, and 8.5 percent of judges
191
False Confession
Prior research contends that false confessions are associated with cases of wrongful
conviction. The following question was asked:
Do you personally know of a case where an innocent individual confessed to a crime he
or she did not commit?
Personal knowledge of a false confession is measured using a Yes-No format. Response
frequencies and percentages are reported in Table 33. Almost 65 percent of all respondents
Table 33
Percentages – False Confessions
Response Police Prosecutors Defense
Attys.
Judges All
1 = Yes 21.4 (59) 17.0 (17) 66.5 (155) 25.4 (45) 35.2 (276)
2 = No 78.4 (214) 83.0 (83) 33.5 (78) 74.6 (132) 64.8 (507)
TOTAL 100.0 (273) 100.0 (100) 100.0 (233) 100.0 (177) 100.0 (783)
had no personal knowledge of a false confession. Conversely, 35 percent of all respondents do
have personal knowledge of a wrongful confession. When group differences are analyzed, the
disparity in the experiences of defense attorneys in comparison to those of the other three groups
is dramatic. Two-thirds of responding defense attorneys personally knew of a false confession,
while less than one-quarter of prosecutors, police, and judges knew of a case where an individual
falsely confessed.
Community Pressure
Community pressure on criminal justice professionals to solve a case has been found in
previous research to be associated with cases of wrongful conviction. To examine this issue one
question is presented.
Do you know of a case in your jurisdiction where community pressure to quickly solve a
crime led to a wrongful conviction?
192
Response frequencies and percentages are reported in Table 34. Personal knowledge of
community pressure is measured using a Yes-No format. Twenty percent of all respondents
Table 34
Percentages – Community Pressure
Response Police Prosecutors Defense
Attys.
Judges All
1 = Yes 3.3 (9) 2.0 (2) 49.8 (116) 15.7 (28) 19.8 (155)
2 = No 96.7 (264) 98.0 (98) 50.2 (117) 84.3 (150) 80.2 (629)
TOTAL 100.0 (273) 100.0 (100) 100.0 (233) 100.0 (178) 100.0 (784)
report personal knowledge of a case of wrongful conviction involving community pressure.
When group differences are analyzed, the disparity in the experiences of defense attorneys in
comparison to the other three groups is, again, dramatic. One-half of responding defense
attorneys personally know of such a case, while only 15.7 percent of judges, 3.3 percent of
police, and 2 percent of prosecutors stated they know of a case involving community pressure
and wrongful conviction. In sum, the percentage of defense attorneys who reported knowing of
a case of wrongful conviction involving community pressure was three times greater than the
percentage reported by judges, fifteen times higher than the percentage reported by police, and
25 times higher than the percentage reported by prosecutors.
Research Question # 6 – What Do Criminal Justice Professionals Believe is an Acceptable
Level of Wrongful Convictions, and Do They Believe Changes are Warranted in the Criminal;
Justice System?
The survey-questionnaire contained two questions regarding what respondents considered
an “acceptable” level of wrongful conviction, and if respondents perceive a need for change in
the criminal justice system.
193
Acceptable Levels of Wrongful Conviction
To measure respondents’ opinions regarding an acceptable level of wrongful conviction a
10-point Likert scale question was presented with values ranging from “0%” to “over 25%.”
Results are reported in table 35. Approximate one-half of all respondents (51.4%) believed that
a “zero” percent rate of wrongful conviction is acceptable. Another one-fourth of all respondents
(26.6%) think that a “less than one-half percent” rate is acceptable, and approximately one-tenth
Table 35
Percentages – Acceptable Level of Wrongful Conviction
Response Police Prosecutors Def. Attys. Judges All
1 = 0% 54.6 (148) 48.5 (48) 49.1 (113) 51.1 (89) 51.4 (398)
2 = < than ½% 24.7 (67) 29.3 (29) 27.0 (62) 27.6 (48) 26.6 (206)
3 = ½ % - 1% 10.7 (29) 13.1 (13) 10.9 (25) 12.6 (22) 11.5 (89)
4 = 1% - 3% 6.3 (17) 8.1 (8) 8.3 (19) 6.3 (11) 7.1 (55)
5 = 4% - 5% 1.1 (3) 1.0 (1) 3.0 (7) 2.3 (4) 1.9 (15)
6 = 6% - 10% 1.5 (4) --- 1.7 (4) --- 1.0 (8)
7 = 11% - 15% 0.7 (2) --- --- --- 0.3 (2)
8 = 16% - 20% 0.4 (1) --- --- --- 0.1 (1)
9 = 21% - 25% --- --- --- --- ---
10 = Over 25% --- --- --- --- ---
TOTAL 100.0 (271) 100.0 (99) 100.0 (230) 100.0 (174) 100.0 (774)
MEAN 1.84 1.84 1.94 1.81 1.86
of all respondents (11.5%) feel that a “one-half to one percent” rate is acceptable. Only ten
percent of all respondents choose a wrongful conviction rate of one percent or more rate as being
an acceptable level. This is the only question in the study of which all groups basically agreed.
An analysis of variance test revealed that no significant differences in the mean scores among the
four groups.
Does wrongful conviction occurs frequently enough to warrant changes in the criminal justice
system?
Respondents’ opinions regarding this question were measured using a Yes-No
format. Approximately two-thirds (63.4%) of all respondents say that changes are not
194
Table 36
Percentages – Changes in Criminal Justice System
Response Police Prosecutors Defense Attys. Judges All
1 = Yes 21.9 (59) 7.1 (7) 74.5 (175) 24.7 (43) 36.6 (284)
2 = No 78.1 (210) 92.9 (92) 25.5 (60) 75.3 (131) 63.4 (493)
TOTAL 100.0 (269) 100.0 (99) 100.0 (235) 100.0 (174) 100.0 (777)
warranted (Table 36). Conversely, approximately one-third of all respondents believe that
changes are warranted. When group responses are compared, most defense attorneys think that
changes are warranted, while only about one-fourth of judges and police believe changes are
warranted. Prosecutors almost unanimously agree (92.9%) that changes in the criminal justice
system are not warranted.
Concluding Question
The final page of the questionnaire provided space for respondents to suggest how the
criminal justice process might be improved. Approximately 60 percent of respondents provided
written suggestions. A sample of these suggestions is presented in chapter fourteen.
195
CHAPTER FOURTEEN
DISCUSSION
The present study examines how frequently systemic errors identified by previous
research occur by asking four groups of criminal justice professionals (police, prosecutors,
defense attorneys, and judges) their opinions regarding this issue. Only one other empirical
study (Rattner 1983) has examined the perceptions of criminal justice professionals regarding the
frequency of wrongful conviction and its’ associated errors. The present study extends this
research by examining six types of police error, five types of prosecutorial error, five types of
defense attorney error, four types of judicial error, and by examining issues related to eyewitness
error, forensic error, false accusations, false confessions, and community pressure.
This chapter contains a discussion of the findings presented in chapter 13. The
discussion is organized around three issues: (1) how frequently do criminal justice professionals
believe certain previously identified types of errors occur, (2) what is the perceived quality of the
work product of each group of professionals, and (3) how frequently do criminal justice
professionals estimate wrongful convictions occur in their own jurisdiction and in the United
States? Each of these issues is discussed in relation to the overall opinions of all respondents to
the survey, and in relation to group differences. The discussion then focuses on placing these
findings in the context of the existing literature, explores implications for policy and future
research, and discusses the limitations of this study.
196
Do criminal justice professionals believe intentional misconduct occurs as often as
unintentional error?
Respondents to the present study believe that intentional error (e.g. intentional eyewitness
misidentification, intentional misrepresentation by forensic experts, prosecutor suppressing
evidence, or police using false evidence) is not a widespread problem in the criminal justice
system, and is not a major factor associated with wrongful conviction. Respondents, in sum,
indicated that ‘intentional’ error occurs very infrequently. There is, however, a belief among
criminal justice professionals that ‘unintentional’ error does occur at various degrees of
frequency. Perceptions regarding how often unintentional error occurs, and under what
circumstances, vary significantly between groups – depending on the category of error.
How frequently do criminal justice professionals believe certain previously identified types
of unintentional errors occur?
Two types of unintentional error are discussed: (1) professional error (error by police,
prosecutor, defense attorney, or judge), and (2) non-professional error (error by eyewitnesses,
forensic experts, false confessions, false accusations, and community pressure).
All Respondents
In the opinion of all respondents, “professional error” is believed to occur most often by
defense attorneys (moderately frequently), and least often by judges (infrequently). Police error
and prosecutor error are viewed as occurring between ‘infrequently’ and ‘moderately frequent’ –
not as often as defense attorney error, but more often than judicial error.
Regarding the five types of “non-professional error” that were a topic of this study,
respondents in total perceive eyewitness error to “seldom” occur, forensic error to occur
197
“infrequently,” and false accusations to occur “occasionally.” Approximately two-thirds
(64.8%) of all respondents know of a case where an innocent person falsely confessed to a crime
he or she did not commit, and about one-fifth (19.8%) of all respondents know of a case where
community pressure to solve a case led to a wrongful conviction.
Prosecutors
Prosecutors, as a group, gave the lowest estimations of error in regard to every topic
area that was a subject of this study. Of the four groups of respondents, prosecutors perceive the
least error across all items. When rating the behavior of their own group, prosecutors attributed
significantly less error to the behavior of their own group than did other groups.
Defense Attorneys
Defense attorneys’ opinions concerning frequency of error were the exact opposite of
those reported by prosecutors. Whereas prosecutors perceive the least frequency of error across
all categories, defense attorneys believe that system error occurs more often, across all
categories, than does any other respondent groups – even when rating errors committed by
defense attorneys.
Police and Judges
In general, police officials and judges agreed with each other. In six out of nine
categories of error (eyewitness error, forensic error, prosecutor error, defense error, judicial
error, false accusations, false confessions) police and judges agreed that error occurred more
often than perceived by prosecutors, but less often than believed by defense attorneys. In two
areas, however, (police error and community pressure) police agreed with prosecutors and not
with judges. In these two areas judges stood alone – perceiving significantly more police error
198
and error due to community pressure than prosecutors or police, but significantly less error than
defense attorneys.
Police and judges, like prosecutors, are conservative in the ranking of their own groups’
error. Both police and judges rate their group’s own rate of error at the lowest level of all four
groups (slightly higher than prosecutors’ estimations, but not significantly different).
What is the perceived quality of the work product of each group of professionals?
All Respondents
In the opinion of all respondents to the survey, the general reliability of judicial decisions
concerning guilt or innocence, and the reliability of evidence presented in court by police
officials and prosecutors was rated as being closer to “usually reliable” than “very reliable.”
Defense attorneys were rated by all respondents to defend their clients closer to “moderately
well” than “very well.”
Respondents to the survey were also asked to comment on the general reliability of
eyewitness and forensic expert testimony. Respondents in total believe the reliability of both
types of testimony to be closer to “usually reliable” than “very reliable” – although when
comparing the reliability of eyewitness testimony to that of forensic expert testimony,
respondents believe that forensic testimony is more reliable than eyewitness testimony.
Prosecutors
Following the pattern set by prosecutors regarding perceptions of error among criminal
justice professionals, prosecutors also estimated the reliability of the work product of criminal
justice professionals to be the highest – in all categories. Prosecutors also give the highest
ratings of the four groups concerning the reliability of eyewitness and forensic testimony.
199
Defense Attorneys
The views of defense attorneys again are in direct contrast with those of prosecutors.
Whereas, among the four groups of professionals, prosecutors rated the highest levels of
reliability across all categories, defense attorneys instead perceive the reliability of the work
product of criminal justice professionals, across all categories, to be the lowest. Defense
attorneys also give the lowest ratings of the four groups concerning the reliability of eyewitness
and forensic testimony.
Police and Judges
Police officials and judges again tend to agree with one another regarding the reliability
of the work product of criminal justice professionals. Members of these two groups generally
agreed that professional outputs were more reliable than believed by defense attorneys, but less
reliable than perceived by prosecutors. Concerning the reliability of eyewitness testimony,
police and judges also agreed that such testimony is more reliable than perceived by defense
attorneys, but less reliable than estimated by prosecutors. These two groups, however, have
different perceptions regarding the reliability of forensic expert testimony. Judges perceive such
testimony is less reliable than believed by police officials.
How frequently do criminal justice professionals perceive wrongful convictions occur in
their own jurisdiction and in the United States?
All Respondents
Respondents to the survey estimate that wrongful conviction occurs in their jurisdictions
in one-half to one-percent of all felony convictions. Respondents further estimate that wrongful
conviction in the United States occurs in between one to three percent of all felony convictions.
200
When asked what an “acceptable” level of wrongful conviction might be – given the reality of
ever present human error – respondents in total agreed that a zero percent rate of wrongful
conviction was the only acceptable level. Finally, when asked if wrongful conviction occurs
frequently enough to warrant changes in the criminal justice process, approximately two-thirds
(63.4%) of all respondents think that changes are not warranted, while approximately one-third
(36.6%) believe that changes are warranted.
Prosecutors
Prosecutors (along with police) perceive the lowest percentage of wrongful conviction in
their own jurisdiction (more than zero but less than one-half percent), and in the United States
(between ½ and 1 percent). Only seven percent of prosecutors thought that the frequency of
wrongful conviction warranted changes in the criminal justice system.
Defense Attorneys
Defense attorneys estimate the highest percentage of wrongful conviction in their own
jurisdictions (between four and five percent) and in the United States (almost 6 percent).
Approximately three quarters (74.5%) of defense attorneys believe that the frequency of
wrongful conviction warrants changes in the criminal justice system.
Police
Police agree with prosecutors that wrongful conviction occurs in their own jurisdictions
more than “never” but in less than one-half percent of all felony convictions, and across the
United States in one-half to one-percent of all felony convictions. Regarding the need for
changes in the criminal justice system, police see more need for change than prosecutors (21.9%
to 7.1%), and less need for change than defense attorneys (21.9% to 74.5%). There were no
201
significant differences in the opinions of police and judges regarding the need for change in the
criminal justice system.
Judges
Judges agree with prosecutors and police that wrongful conviction occurs in their own
jurisdiction in less than one-half percent of cases, but although all three groups selected the “less
than one-half percent” range, judge’s estimates are significantly higher than those of prosecutors
and police (i.e. closer to one-half percent). In regard to the frequency of wrongful conviction
across the United States, judges perceive a higher frequency rate (between one and three percent)
than do prosecutors and judges, but a lower rate than estimated by defense attorneys. Finally, in
regard to the need for changes in the criminal justice system, judges agree with police.
Significantly more judges and police believe that changes are warranted than do prosecutors, and
significantly less believe changes are warranted than do defense attorneys
Overall Conclusions
Certain trends are apparent in the findings. Across all categories, in regard to error,
reliability of work product, and frequency of wrongful conviction, prosecutors proffered the
lowest estimates of error. Their views appear to be “defending the system.” Prosecutors, in sum,
perceive little error, high reliability, and a low frequency of wrongful conviction. Defense
attorneys are at the opposite pole. Their views appear to be more critical of the criminal justice
system. In contrast to prosecutors, defense attorneys perceive the highest level of error of all
groups, the lowest level of reliability, and the highest estimates concerning the frequency of
wrongful conviction. Judges and police tend to agree with each other that the “system” is not as
202
error-prone or unreliable as perceived by defense attorneys, but more prone to error and less
reliable than is estimated by prosecutors.
A point needs to be made concerning the issue of “frequency” of error. Several types of
error are discussed in this study. The frequency of error is often perceived by respondents to
“seldom” occur. Likewise, evidence presented in court by prosecutors and police, or judicial
decisions are often believed to be “usually” reliable. On the surface these ratings may appear
acceptable. However, in the case of the criminal justice system, error that occurs “seldom,” or
evidence that is “usually” reliable – may be indications that error is occurring too often – at least
from the point of those who would like the criminal justice system to function more effectively.
In many areas of public administration “seldom” would be acceptable, e.g. when power failures
seldom occur, or when planned meetings are seldom missed. In the field of criminal justice,
however, errors that “seldom” occur can have devastating, long-lasting, and even fatal
consequences to the lives of numerous people, e.g., the wrongfully convicted themselves, their
families, trial witnesses, jurors, and even criminal justice professionals themselves.
Respondents to the survey were given a range of possible responses for each question in
the survey – usually from “never” to “always,” or from “very reliable” to “very unreliable.” It
was not expected that respondents would indicate that error “always” occurs. Likewise, it was
not expected that most respondents would believe that error “never” occurs. It was expected that
respondents would perceive that most types of system error occurs “infrequently” or less. What
in fact occurred was, when respondents were given an opportunity to rate the frequency of
specific error, they rated error as occurring between “infrequent” and “moderately frequent,”
instead of between “never” and “infrequent.” Likewise respondents had opportunities to rate the
reliability of police and prosecutorial evidence, and the reliability of judicial decisions, as being
203
closer to ‘very reliable’ than ‘usually reliable,’ but instead they rated reliability closer to ‘usually
reliable.’ Finally, respondents had opportunities to rate how well defense attorneys defend their
clients as closer to “very well” than “moderately well,” but instead they rated it closer to
“moderately well.” The results of this survey, therefore, indicate unacceptably high rates of
professional error and unacceptably low rates of professional reliability.
A similar point can be made regarding the “frequency” of wrongful conviction per se. A
significant finding of this study is that, when the views of all respondents are tallied, there is a
belief that wrongful conviction occurs at a rate of between one-half and one percent in
respondents’ jurisdictions, and at a rate of between one and three percent across the United
States. Today there are in excess of 2,000,000 individuals incarcerated in our nation’s prisons
and jails (Silverman, 2001). If error “only” leads to a wrongful conviction in one percent of all
cases, there are then presently 20,000 individuals currently wrongfully incarcerated in the United
States – some possibly on death row. If the three percent estimate is correct, there are 60,000
individuals currently serving jail and prison sentences for crimes they did not commit. In other
words, according to the criminal justice professionals that were the subject of this study, there
are currently between 20,000 and 60,000 individuals incarcerated in the United States for crimes
they did not commit.
The consequences of error in the criminal justice system might best be compared to the
consequences of error in the commercial airline industry. During the year 2000 there were 18
million commercial flights with 20 fatal accidents – a .0000011 “error rate” (Byrne-Crangle,
2001). In contrast, in the year 2000 there were 2,000,000 prisoners incarcerated in U.S. prisons
and jails with between 20,000 and 60,000 of those individuals wrongfully convicted – an error
rate of 1 to 3 percent. If an error rate of 1 to 3 percent was experienced by the airline industry
204
there would be between 180,000 and 540,000 fatal commercial airline accidents per year. To put
it another way, if commercial airlines experienced the same error rate as does the criminal justice
system, between one and three of every 100 commercial airlines that took off would crash.
Findings in the Context of Existing Literature
The present findings confirm many of the findings of Scheck, Neufield, and Dwyer
(2000), as well as the findings of the National Institute of Justice study on wrongful conviction
(Conners, Lundregan, Miller, and McEwen 1996). These studies suggest that certain types of
error are associated with the incidence of wrongful conviction, and that wrongful conviction per
se does occur, albeit relatively infrequently, but in sufficient quantity to disrupt the
administration of justice in the United States and to negatively affect the lives of thousands of
individuals. The Scheck et al. (2000) study, however, suggests higher levels of professional
misconduct and intentional error than was found in the present study.
The present study also confirms and expands on Rattner’s (1983) pioneering study of the
perceptions of criminal justice professionals concerning the topic of wrongful conviction. These
new data are important because: (1) they confirm Rattner’s findings that wrongful conviction
occurs more often than is believed by some individuals – hopefully leading to a greater
“consciousness” of the problem by policymakers and criminal justice professionals, and (2) they
expand Rattner’s study by investigating areas of potential error that were not a subject of his
research – hopefully leading to a better “understanding” of the factors associated with the
phenomenon, and ultimately to solutions that will reduce the incidence of wrongful conviction.
In the present study almost four out of five respondents (79.5%) reported that wrongful
conviction occurs in their own jurisdiction. This figure is higher than was reported by Rattner
205
(1983) almost twenty years ago in his survey of a similar group of criminal justice professionals
when only 63.3% of respondents reported that wrongful conviction occurred in their
jurisdictions. It should be noted that Rattner’s study took place before modern DNA
identification methods were able to be used – methods which have ultimately led to the
exoneration of numerous wrongfully convicted individuals. It is likely that knowledge of these
exonerations has increased the awareness of criminal justice professionals to the problem of
wrongful conviction and influenced their perceptions as to the frequency of the phenomenon.
Also, in the present study, the frequency of wrongful conviction in respondents’ own
jurisdictions is estimated at between one-half and 1 percent of all felony cases. Respondents to
Rattner’s survey estimated the level of wrongful felony conviction in their own jurisdictions to
be “less than one percent” – a finding that appears to be similar to that of the present study.
However, in the Rattner study, 36.8 percent of respondents believed that wrongful felony
conviction “never” occurred in their own jurisdiction, while in the present study only 19.5
percent of respondents believe wrongful conviction “never” occurs in their jurisdictions.
When asked to estimate how frequently wrongful conviction occurs in the United States,
respondents to Rattner’s survey estimated a rate of “less than one percent” – the same as they
had estimated for their own jurisdictions. In the present study, respondents estimate a national
rate of between 1 and 3 percent of all felony convictions.1 Almost six percent (5.6%) of those
professionals responding to Rattner’s survey believed wrongful felony conviction “never”
occurred, while only seven-tenths of a percent (0.7%) of respondents to the present survey
believed wrongful felony conviction “never” happens. The largest contrast in estimations
1 This estimate is higher than estimates by respondents for their own jurisdictions. It is believed the
differences in these two estimates result from estimations made by respondents from smaller jurisdictions,
where smaller caseloads prevail, and rates of wrongful conviction may be lower than in larger jurisdiction
where caseloads are heavier.
206
between the two studies is indicated in the ‘one to five percent’ category. Six and one-half
percent of respondents to Rattner’s survey estimated that wrongful conviction occurs in 1 to 5
percent of all felony convictions. In contrast, almost four times as many respondents to the
present survey (23.1 %) estimate that wrongful conviction occurs in 1 to 5 percent of all felony
cases. In sum, when the results of the current study are compared with the Rattner study of 20
years ago, perceptions and estimations of wrongful conviction appear to have increased.
Whether these increases are due to factors associated with the revelations of modern DNA
technology, and the accompanying media attention and professional discussion, remains an open
empirical question.
Policy Implications
One finding of this study is that there is a gap between what the professionals in the
criminal justice system think is an acceptable level of wrongful convictions in the United States,
and what they perceive to be the current level. When respondents were specifically asked what
they considered to be an acceptable level of inevitable wrongful conviction, the majority (51.4%)
declared that only a zero level of wrongful conviction is acceptable; when asked what they
believe is the current rate of wrongful conviction, a majority of respondents (64.7%) perceive a
one to three percent (or greater) error rate. A foundation for reform, therefore, appears to be in
place.
According to the respondents of this survey, reductions in the rate of system error are
possible. The final two pages of the survey instrument used to conduct this study provided space
that allowed respondents to offer suggestions regarding how, in their professional opinions, the
incidence of wrongful conviction might be reduced. One judge wrote:
The tools are available to improve upon reducing the rate of
wrongful conviction. However, the overwhelming volume
207
of cases imposed upon the courts in certain jurisdictions
within certain time spans without adequate investigation
and/or preparation by law enforcement, prosecutors,
defense counsel and the courts, will always impact upon a
perfect system. Leadership by professionals in the
respective fields will affect the wrongful conviction rate.
Two-thirds of the 698 respondents offered written comments. Respondents to the
survey had numerous insights and suggestions to policymakers as to how the system could be
improved to reduce error.
Unintentional Error
One policy implication of this research is that resources targeted at reducing the
incidence of wrongful conviction should be targeted at reducing “unintentional” error. The
findings of this study clearly indicate that professional misconduct and intentional error –
although disturbing and unacceptable at any level – are not viewed as significant a factor
associated with wrongful conviction as is unintentional error. Whatever initial resources that
may be allocated to reduce the incidence of wrongful conviction would therefore best be spent
implementing policies or programs to reduce unintentional error.
Reduce System “Overload”
Members of all groups agree that many actors in the criminal justice system are
overwhelmed by the avalanche of cases – especially in large cities. Over 100 respondents
provided written suggestions and comments concerning high caseloads which may contribute to
mistakes that could lead to wrongful conviction. A chief county prosecutor wrote:
The biggest problem I see is that all of us – at every step –
have so darn many cases that it is sometimes very difficult
to do the job you know needs to be done.
A judge wrote:
Prosecutors need to focus on more important cases, by more
208
intense investigation and trial preparation. Many are
overburdened by lesser felony cases that they are not
prepared for trial, thus – errors on the most serious felonies.
A second policy implication of this study would therefore be that caseloads be reduced by either:
(a) increasing personnel, or by (b) narrowing the “net-of-law.”
Increased Professionalism
The findings of this study also suggest that policies be devised to increase
professionalism in the criminal justice system. Almost 100 respondents provided written
comments suggesting that wrongful convictions might be reduced if more stringent selection and
hiring standards were put in place in order to reduce the number of individuals in the criminal
justice system that are ill-suited for their jobs and who often approach their duties with biases
and ill-conceived ideas of justice. Another consistent suggestion was that better training should
be provided (including ethical training). Training should be enhanced both when individuals are
initially hired, and then on a continuing basis. Finally, many respondents call for higher
professional standards and stricter discipline for those who engage in professional misconduct
and for those criminal justice actors who consistently make unintentional errors. A police
official wrote:
Set minimum requirements of training and continuing legal
education for criminal defense attorneys – particularly in
death-penalty cases nationwide. Also, a minimum
requirement of continuing training hours (annually) for
police officers in general, and detectives in particular,
regarding identification procedures, forensic evidence
collection, and securing statements, in addition to others.
Others have suggested that improved selection and training policies would improve
professionalism, and reduce professional misconduct and unintentional error (Deakin, 2001;
Kelly and Wearne 1998; Sammons 1988; Wroblesky and Hess 2003). Concerning police
209
training, Birzer and Tennehill (2001:233) note, “there is an obvious need for police officers to
acquire knowledge of the latest legal decisions, technological advances, and tactical
developments in the field, and to remain proficient in a number of job-related skills.”
In addition to overall concerns regarding the overburdened criminal justice system and
the need for professionalism, there were numerous suggestions with policy implications targeted
at each particular group that was a part of this study.
Police Practices
Policy implications for police practices were suggested by numerous respondents. A
major topic concerning the police involved criticisms of the lack of thoroughness of many police
investigations. One police official said:
Law enforcement investigators absolutely need to build
their investigation on factual information only. Many
experienced and inexperienced investigators form an
opinion at the onset of their investigation and develop their
fact base on their preconceived opinions. The facts need to
speak for themselves.
Police practices involving identification procedures were also a topic of much criticism.
The majority of suggestions revolved around the need to videotape all police interrogations and
confessions. A defense attorney wrote:
Supposed “confessions” are often summaries prepared by the
police and signed by an undereducated, unsophisticated,
client who is assured of leniency for cooperating and giving a
statement. In this day and age, all statements should be
videotaped. In fact, arrest to booking should be videotaped.
When police video or audio an interview, they frequently do
a “run-through” ahead of time. What is not on the tape is the
three hours of interviews where the accused is told he “will
fry” if he doesn’t confess or some other such nonsense
Prosecutorial Practices
210
Prosecutors also received their share of criticism – primarily revolving around
overzealousness and not providing open discovery. Regarding overzealousness, one defense
attorney commented, “Chief prosecutor should refrain from pushing their assistant ‘to win at all
costs’ irregardless of justice,” while a judge wrote, “Prosecutors need to be better prepared, less
over-indicting, and realize the abuses of their power and be held accountable.” Another typical
comment came from a police official who said, “Prosecuting attorneys are “graded” on
conviction rates. Young prosecutors appear pressured to get some type of conviction. Emphasis
must be placed on the truth.”
Regarding the need for prosecutors to provide “open” discovery, a criminal defense
attorney wrote:
The rules of discovery are an embarrassment to the right of
a defendant to a fair trial. The defendant is forced to
blindly go to trial with a small fraction of disclosed
evidence. The Ohio Rules of Criminal Procedure is the
most significant source of wrongful conviction, followed
closely by overzealous police and prosecutors.
Defense Attorney Practices
Defense attorneys were often criticized for lack of preparedness. A police official wrote,
“I feel public defenders do not prepare their cases adequately,” and a judge commented,
“Improve the quality of the defense bar.” A major topic concerning the ability of defense
attorneys to properly defend their clients revolved around the need for adequate resources to
combat the resources available to the prosecutor. A prosecutor commented:
Police officers do occasionally get sloppy and arrest the
wrong person. It would be nice if we could make police
more through and objective, but the reality is that our
system relies on defense lawyers to make sure justice is
done. Unfortunately it is rare for a lawyer to have the
211
resources necessary to properly investigate and defend. In
other words, sadly, justice requires money.
Judicial Practices
Criticism of judges primarily revolved around personal bias and judicial pressure exerted
on the plea-bargaining process. One defense attorney wrote:
Most judges in my county are less than impartial in
criminal cases. Individuals very often confess to a crime
they did not commit because most judges will tell the
defense attorney they will max your client’s sentence if
they go to trial and lose.
Increased Diligence
Reducing the incidence of wrongful conviction likely involves a more intense study of
why and how the phenomenon occurs than is currently being undertaken. Although there are
presently more researchers looking at the phenomenon than ever before, many policymakers who
might be in the position to make warranted changes are not doing so. Of course no one today
wants to appear “soft-on-crime.” There is also present the belief among many in the criminal
justice system that, if additional resources are to be spent, they should be targeted at increasing
the conviction rate instead of the exoneration rate. Comments volunteered by respondents to this
study suggest the controversial nature of research on wrongful conviction. One respondent
commented, “Why don’t you study wrongful acquittal.” Another wrote, “It would be nice, for
once, if academics such as yourself could initiate research projects that actually help those of us
engaged in the everyday work of the criminal justice system.” Of course there were also
encouraging words. One police officer stated, “Hopefully I have never convicted an innocent
person. I have never perjured myself in court, even if it meant losing the case. I would rather
see a 1000 guilty people go free than convict an innocent person,” and a judge wrote, “Our
212
forefathers intended the criminal justice process to be difficult so innocent persons would not be
prosecuted. The law should always be a check or a governor on the police power of the state.
Thank you for your vigilance.”
Just as the commercial airline industry dissects each airline accident, the criminal justice
system should likewise dissect each proven wrongful conviction in order to determine how it
occurred and what can be done to prevent, or at least reduce, the chances that such errors
reoccur. The goal should be to reduce the error rate in the criminal justice process from its
current .01 to .03 rate to the commercial airlines’ error rate of .0000011. This will only be
possible if those in the criminal justice system give the same intense attention to the types of
error that are the subject of this study as do the overseers of the commercial airline industry give
intense attention to the causes of airline crashes. Each proven case of wrongful conviction
should be placed in a national registry, and then studied in order to determine when, where, why,
and how the criminal justice process faltered.
Directions for Future Research
Prior research has only indicated that the types of errors that were a subject of this study
are “associated” with wrongful conviction. To-date, no direct causal relationship has been
proven. Future research should focus on determining the causal elements and relationships that
lead to wrongful conviction – of which there may be many. The present study, as well as all
previous research, suggests that no single factor “causes” a wrongful conviction. In almost every
proven case of wrongful conviction, numerous errors had occurred in the processing of the
defendant. The challenge for the future, therefore, is to determine which errors, and which
interactions contribute to wrongful conviction, and to make that information available to policy
makers so remedial steps may be taken.
213
Future research regarding the perceptions of criminal justice professionals should be
conducted in other geographic areas to determine the generalizability of the findings of the
present study (which was limited to one geographic area – the State of Ohio). Finally, future
research should be conducted to determine if the dynamics of wrongful conviction vary when
large metropolitan jurisdictions are compared to smaller rural jurisdictions.
214
REFERENCES
American Bar Association. 1979. Annotated Code of Professional Responsibility. Chicago:
American Bar Association.
ABA Journal. March, 1993.
ABA Journal. Feb. 1996.
ABA Journal. July 1989. “Crime and Punishment”
Adams, R.D., with W. Hoffer and M. Hoffer (1991). Adams v. Texas. New York: St. Martin’s.
Allport, G.W. and L.J. Postman. 1974. The Psychology of Rumor. New York: Henry Holt.
American Bar Association. 1979. Model Code of Professional Responsibility and Code of
Judicial Conduct. Chicago: American Bar Association.
Arenella, P. 1968. “The Prosecutor’s Role in Plea Bargaining.” University of Chicago Law
Review. 36:50, pp 62-3.
Arnold, T.W., Kamisar, Y. and F.E. Inbau. 1965. Criminal Justice in Our Time. Charlottesville:
University Press of Virginia.
A Second Rampart Officer Tells of Corruption. 2000 (January 28). Los Angeles Times.
Bandes, S. 1996. “Simple Murder: A Comment on the Legality of Executing the Innocent.”
Buffalo Law Review 44:501-25.
Bartz, A.E. 1999. Basic Statistical Concepts. 4th. New Jersey: Prenice-Hall.
Becker, H.S. 1963. Outsiders: Studies in the Sociology of Deviance. New York: The Free Press.
Bedau, H.A. and M.L. Radelet. 1987. “Miscarriages of Justice in Potentially Capital Cases.”
Stanford Law Review 40: 21-179.
Bedau, H.A. and M.L. Radelet. 1988. “The Myth of Infallibility: A Reply to Markman and
Cassell.” Stanford Law Review 41:161-70.
Bentham, J. Principles of Penal Law, 1 Works 476 in J. Browning ed. 1962.
Blumberg, A. 1969. “The Practice of Law as a Confidence Game.” Sociology of Law 12:
321-331.
Blumstein, A. 1998. “U.S. Criminal Justice Conundrum: Rising Prison Populations and Stable
Crime Rates.” Crime and Delinquency Vol. 44: 127-135.
Borchard, E.M. 1932. Convicting the Innocent: Sixty-Five Actual Errors of Criminal Justice.
Garden City, N.Y.: Doubleday.
Brandon, R. and C. Davies. 1973. Wrong Imprisonment: Mistaken Convictions and Their
Consequences. London: Allen & Unwin.
Break Rules, Be Promoted. 1999 (January 14). Chicago Tribune.
Brigham, J.C. and P. Barkowitz. 1978. “Do they all look alike? The effects of race, sex,
experience, and attitudes on the ability to recognize faces.” Journal of Applied
Psychology 8:306-18.
Bright, S.B. 1994. “Counsel for the Poor: The Death Sentence Not for the Worse Crime but for
the Worse Lawyer.” The Yale Law Journal 103: 1835-83.
Block, E. 1963. The Vindicators. Garden City, NY: Doubleday.
Brunk, C. 1979. “The problem of Voluntariness and Coercion in the Negotiated Plea.” Law and
Society 13: 522-52.
Buckhout, R. 1974. “Eyewitness testimony.” Scientific American 231: 23-31.
Buckholt, R, D. Figueroa, and E. Hoff. 1975. “Eyewitness Identification: Effects of suggestion
and bias in identification from photographs.” Bulletin of Psychonomic Society 6:71-4.
215
Buckhout, R. 1977. “Eyewitness identification and psychology in the courtroom.” Criminal
Defense 4:5-10.
Buckhold, R.D. and M. Greenwald. 1980. “Witness Psychology.” In J. Imwinkelreid (ed.),
Handbook of Forensic Evidence. New York: Practicing Law Institute.
BJS. 1995. Bulletin. Washington DC: U.S. Government Printing Office, April 1995, p. 13.
BJS. 1998. Nation’s Probation and Parole Population Reached New High Last Year, BJS News
Release. Washington DC: August 16, 1998.
BJS. 1998. Bulletin. August 1998. p.5.
Byers, B. 2000. “Ethic and Criminal Justice: Some Observations on Police Misconduct.” Justice
Quarterly. 21,3: 1.
BJS. 1999. Criminal Offender Statistics, at http.//www.ojp.usdoj.gov/bjs/crimoff.htm (revised
12/5/99).
Castelle. G. 1999. “From the ‘Fred Zain Affair.” The Champion 23,4: 13-14; 52-57.
Ceci, S.J., M. Toglia and D. Ross (Eds.) 1987. Children’s Eyewitness Memory New York:
Springer-Verlag.
Christenson, S.A. 1992. “Emotional Stress and Eyewitness Testimony: A Critical Review.”
Psychological Bulletin, 112:284-309.
Church, T. 1979. “In Defense of Bargain Justice.” Law and Society 13: 509-25.
Clear, T.R. and G.F. Cole. 2000. American Corrections. Belmont: Wadsworth
Clear, T.R. 1994. Harm in American Penology: Offenders, Victims, and Their Communities.
Albany:State University of New York Press.
Clendenning, A. “Test Frees Man Jailed For Rape.” Associated Press. November 26, 2000.
Clifford, B. and J. Scott. 1978. “Individual and situational factors in eyewitness testimony.”
Journal of Applied Psychology 63: 352-54.
Connery, D. 1996 ed. Convicting the Innocent: The Story of a Murder, a False Confession and
the Struggle to Free a ‘Wrong man’. Cambridge, MA: Brookline Books.
Connors, E. T. Lundregan, N. Miller, and T. McEwen. 1996. Convicted by juries, exonerated by
science: Case studies in the use of DNA evidence to establish innocence after trial.
Alexandria, VA: National Institute of Justice.
Convictions of 9 More Voided in Scandal. 2000 (February 2). Los Angeles Times.
Cooper, C., Cooper S. and S. Reese. 1995. Mockery of Justice: True Story of the Sam Shepard
Murder Case. Boston: Northeastern University Press.
Coursey, D. and L. Stanley. 1988. “Pretrial Bargaining Behavior Within the Shadow of the
Law: Theory and Experimental Evidence. International Review of Law and Economics.
8: 161-79.
Court stand-in is convicted of crime he didn’t commit. July, 17, 1980, p. 12. Atlanta
Constitution.
Crank, J.P. and M.A. Caldero. 2000. Police Ethics: The Corruption of Noble Cause. Cincinnati:
Anderson Publishing Co.
Crime and Punishment in Dallas. (1989 July). ABA Journal, pp. 52-56.
Cullen, F.T. and K.E. Gilbert. 1982. Reaffirming Rehabilitation. Cincinnati: Anderson
Publishing Company.
Currie, E. 1985. Confronting Crime: An American Challenge. New York: Pantheon Books.
Cutler, B.L. and S.D. Penrod. 1988. “Improving the reliability of eyewitness identification: Lineup
construction and presentation.” Journal of Applied Psychology 73: 281-90.
216
Cutler, B.L. Penrod, S.D. and T.K. Martens. 1987. “The reliability of eyewitness identifications:
The role of system and estimator variables.” Law and Human Behavior 11: 223-58.
Cutler, B.L., S. Penrod and T. Stuve. 1988. “Jury Decision Making in Eyewitness Cases.” Law
and Human Behavior 12:41-56.
Cutler, B.R., S. Penrod and H. Dexter. 1990. “Juror sensitivity to eyewitness identification
evidence.” Law and Human Behavior 14:185-91.
Currie, E. 1998. Crime and Punishment in America. New York: Metropolitan Books.
Davis, N. 1991. White Lies: Rape, Murder, and Justice Texas Style. New York: Pantheon
Death Penalty Information Center. Innocence and the Death Penalty: The Increasing Danger of
Executing the Innocent. Washington D.C. July 1997 (with February 24, 1998 update).
Dekle, D.J., Beale, C.R., Elliot R., and D. Huneycutt. 1996. “Children as witnesses: A
comparison of lineup versus showup methods.” Applied Cognitive Psychology 10:1-12.
Dieter, R.C. 1997. “Innocence and the Death Penalty: The Increasing Danger of Executing the
Innocent.” Death Penalty Information Center. July 1997.
Dillman, D.A. 2000. Mail and Internet Surveys: The Tailored Design Method. 2nd. New York:
John Wiley & Sons, Inc.
Dershowitz, A. 1982. The Best Defense. New York: Random House.
Devenport, J.L., S. Penrod and B. Cutler. 1998. “Eyewitness identification evidence: Evaluating
commonsense evaluations.” Psychology, Public Policy, and Law. 3: 338-61.
Fanselow, M.S. and R.F. Buckhout. 1976. Nonverbal cueing as a source of biasing information
in eyewitness identification testing. New York: Center for Responsive Psychology,
Brooklyn College C.U.N.Y.
Feeley, M. and J. Simon. 1992. “The New Penology: Notes on the Emerging Strategy of
Corrections and its Implications.” Criminology 30.4: 449-74.
Fisher, R.P. 1995. “Interviewing Victims and Witnesses of Crimes.” Psychology, Public Policy,
and Law, 1:732-64.
Four More LAPD Officers Are Suspended. 2000 (January 14). Los Angeles Times.
Four Officers Face Charges. 2000 (November 3). The Cincinnati Enquirer.
Frank, J. and B. Frank. 1957. Not Guilty. New York: Doubleday.
Friedman, L. 1979. “Plea Bargaining in Historical Perspective.” Law and Society 13: 247-59.
Frisbie, T. and R. Garrett. Victims of Justice. NY: Avon.
Gardner, E.S. 1952. Court of Last Resort. New York: William Sloane.
Gershman, B. 1999. Prosecutorial Misconduct. Deerfield, Ill.: Clark Boardman Callaghan.
Gordon, D. 1991. The Justice Juggernaut: Fighting Street Crime, Controlling Citizens. New
Brunswick: Rutgers University Press.
Granelli, J.S. 1980. “Trials – and errors.” National Law Journal. December 15, 1980.
Gregory, W.L., J. C. Mowen and D.E. Linder. 1978. “Social Psychology and Plea Bargaining:
Applications, Methodology and Theory. Journal of Personality and Social Psychology.
36:1521-30.
Gross, S.R. 1996. “The Risks of Death: Why Erroneous Convictions Are Common in Capital
Cases.” Buffalo Law Review 44:469-500.
Grumbaum, B. 1985. “Crime Labs Not Reliable, Criminalist Claims.” Chemical and
Engineering News. May 20, 1985, p. 40.
Heumann, M. 1978. Plea Bargaining: The Experience of Prosecutors, Judges, and Defense
Attorneys Chicago: University of Chicago Press.
Hirsch, J. 2000. The Miraculous Journey of Rubin Carter. NY: Houghton Muffin.
217
Hosch, H. 1994. Individual differences in personality and eyewitness identification. In D.F.
Ross, J. D. Read, and M.P. Toglia (Eds.), Adult Eyewitness Testimony: Current Trends and
Developments (p. 328-47). New York: Cambridge University Press.
Huff, R.C. and A. Rattner. 1988. “Convicted But Innocent: False Positives and the Criminal
Justice Process,” in Joseph E. Scott and Travis Hirschi, eds., Controversial Issues in
Crime and Justice. Newbury Park, CA: Sage.
Huff, C.R., A. Rattner and E. Sagarin. 1996. Convicted but Innocent: Wrongful Conviction and
Public Policy. London: Sage Publications.
Huff, C.R., A. Rattner and E. Sagarin. 1986. “Guilty Until Proven Innocent: Wrongful
Conviction and Public Policy.” Crime and Delinquency 32: 518-44.
Humes, E. (1999). Mean Justice: A Town’s Terror, A Prosecutor’s Power, A Betrayal of
Innocence. New York: Simon & Schuster Inc.
Inbau, F.E. et al. 1986. Criminal Interrogation and Confessions.
Katz, B.S. 1997. Justice Overruled: Unmasking the Criminal Justice System. N.Y.: Warner
Books.
Kassin, S.M., Ellsworth, P., and V.L. Smith. 1989. “The ‘general acceptance’ of psychological
research on eyewitness testimony.” American Psychologist, 44: 1089-98.
Kassin, S.M. 1997. “The Psychology of Confession Evidence.” 52 American Psychologist 221.
Kingsnorth, R.F., MacIntosh, R.C. and J. Wentworth. 1999. “Sexual Assault: The Role of Prior
Relationship and Victim Characteristics in Case Processing.
Klockers, C.B. 1980. “The Dirty Harry Problem.” The Annals, 452: 33-47.
Klockers, C.B. 1991. “The Rhetoric of Community Policing.” p.p. 530-42 in C. Klockers and S.
Mastrofski (eds.) Thinking About Policing, Second Edition, NY: McGraw Hill.
LAPD Chief Call for Dismissal of All Rampart Cases. 2000, January 27). Los Angeles Times.
LAPD Corruption Probe Grows in 7 Shootings, New Allegations. 1999 (October 22). Los
Angeles Times.
Lavakas, P.J. and L. Bickman. 1975. What Makes a Good Witness? Paper presented at the annual
meeting of the American Psychological Association, Chicago.
Lane, M. 1970. Arcadia. New York: Holt, Rinehart and Winston.
Law Dictionary, The (Gilmer’s Revision). 1986. Gilmer Jr., W. Cincinnati, Ohio: Anderson
Publishing Company.
Leo, R.A., and R.J. Ofshe. 1998. “The Consequences of False Confessions: Deprivations of
Liberty and Miscarriages of Justice in the Age of Psychological Interrogation.” Journal of
Criminal Law and Criminology. 88 (2): 429-96.
Lindsay, R.C. and J.D. Read. 1995. “Memory Work – and recovered memories of childhood
sexual abuse: scientific evidence and public, professional, and personal issues.”
Psychology, Public Policy, and Law. 1: 846-909.
Lindsay, R.C., Wallbridge, H., and D. Drennan. 1987. “Do clothes make the man? An
exploration of the effect of lineup attire on eyewitness identification accuracy.” Canadian
Journal of Behavioral Science 19:463-78.
Lindsay, R.C. and G.L. Wells. 1980. “What Price Justice? Exploring the relationship between
lineup fairness and identification accuracy.” Law and Human Behavior. 4: 303-14.
Lindsay, R.C. and G.L. Wells. 1985. “Improving eyewitness identifications from line-ups:
Simultaneous versus sequential line-up presentations.” Journal of Applied Psychology 70:
556-64.
218
Lindsay, R.C., Lea, J.A., and J.A. Fulford. 1991. “Sequential line-up presentation: Technique
matters.” Journal of Applied Psychology 76: 741-75.
Linscott, S. and R.L. Frame. 1994. Maximum Security. Wheaton Illinois: Crossway Books.
Loftus, E.F. 1979. Eyewitness Testimony. Cambridge, MA: Harvard University Press.
Loftus, E.F. and K. Ketcham. 1994. The Myth of Repressed Memory: False Memories and
Allegations of Sexual Abuse. New York: St. Martin’s Press.
Luus, C.A. and G.L. Wells. 1994. “The malleability of eyewitness confidence: Co-witness and
perseverance effects.” Journal of Applied Psychology. 79:714-24.
Lykken, D. 1981. “The Truth About Lie Detectors.” People Weekly May 11, 1981, p.45.
MacNamara, D.E.J. 1969. “Convicting the innocent.” Crime and Delinquency 15:57-61.
McCloskey, J. 1996. “The Death Penalty: A Personal View.”Criminal Justice Ethics 8/1: 2, 54-9.
McCloskey, J. 1989. “Convicting the Innocent.” Criminal Justice Ethics. 8-1: 2; 54-59.
McMahon, M. 1995. “False Confessions and Police Deception: The Interrogation, Incarceration,
and Release of An Innocent Veteran” 13 Am J. Forensic Psychologist 5.
Malpass, R.S. and P.G. Devine. 1981. “Eyewitness identification: Lineup instructions and the
absence of the offender.” Journal of applied Psychology 66:482-89.
Malpass, R.S. and J. Kravitz. 1969. “Recognition of faces of own and other races.” Journal of
Personality and Social Psychology 13:330-34.
Man Revels in His Freedom. 1989 (April 27). Columbus Dispatch.
Markman, S.J. and P.G. Cassell. 1988. “Protecting the Innocent: A Response to the Bedau-
Radelet Study.” Stanford Law Review 41: 121-60.
Martinson, R. 1974. “What Works? – Questions and Answers About Prison Reform.” Public
Interest (Spring): 22-54.
Maxfield M. G. and E. Babbie. 1998. Research Methods for Criminal Justice and Criminology.
Cincinnati: Wadsworth Publishing Company.
Mauer, M. 1995. America Behind Bars: U.S. and International Use of Incarceration.
Washington, DC: The Sentencing Project, 1997.
Mnookin and Kornhauser (See Prosecutor – plea bargaining)
Morris, E. (Director), (1988). The thin blue line [Film]. New York: Miramax Films.
Munsterberg, H. 1908. On the Witness Stand: Essays on Psychology and Crime. Berlin: Fred B.
Rothman & Co.
National Law Journal. 1999. A Good Year for Those Who Fight To Save the Innocent From
Death. In Criminal Justice Weekly Volume 2, No. 1, p 24-5
New York Times. Feb 4, 1997.
New York Times. October 29, 1997.
New York Times Magazine. December 11, 1983, p. 103. “Experts’ Day In Court.”
Ohio Native Finally Cleared in Florida. 1990 (September 5). Columbus Dispatch.
Ofshe, R. J. and R. A. Leo. 1997. “The Decision to Confess Falsely: Rational Choice and
Irrational Action.” 74 DENV. U.L. Rev. 979
Packer. H.L. 1968. The Limits of the Criminal Sanction. New York: Henry Holt.
Parker, J.F. and L.E. Caranza. 1989. “Eyewitness testimony of children in target-present and
target-absent lineups.” Law and Human Behavior. 13:133-49.
Parker, J.F., Haverfield, E., and S. Baker-Thomas. 1986. “Eyewitness testimony of children.
Journal of Applied Social Psychology 16:287-302.
Potter, J. and F. Bost. 1997. Fatal Justice: Reinvestigating the Mac Donald Murders. NY:
Norton.
219
Protess, D., Warden, R., and R. Warden. 1998. A Promise of Justice: The Eighteen-Year Fight
to Save Four Innocent Men. New York: Hyperion.
Radelet, M.L., Bedau, H.A., and C.E. Putnam. 1992. In Spite of Innocence: Erroneous
Convictions in Capital Cases. Northeastern University Press.
Radelet, M.L., Lofquist, W.S., and H.A. Bedau. 1996. “Prisoners Released From Death Row
Since 1970 Because Of Doubts About Their Guilt.” Thomas M. Cooley Law Review
13:907-66.
Radin, E.D. 1964. The Innocents. New York: William Morrow
Rattner, A. 1983. When Justice Goes Wrong: Convicting the Innocent. Unpublished doctoral
dissertation, Ohio State University.
Reiman, J. 2001. The Rich Get Richer and the Poor Get Prison: Ideology, Class, and Criminal
Justice. Boston: Allyn and Bacon.
Rampart Probe May Now Affect Over 3,000 Cases. 1999. (December 15). Los Angeles Times.
Ramsey, R.J. 1994. A Failure of the Lawyer Discipline System: The Need for an Independent
Agency to Regulate Discipline in the Legal Profession. Unpublished master’s thesis,
University of Cincinnati.
Saks, M. J. 1998. “Merlin and Solomon: Lessons from the law’s formative encounters with
forensic identification science. Hastings Law Journal 49: 1069.
Saks, M. J. and J.J. Koehler. 1991. “What DNA ‘fingerprinting’ can teach the law about the rest
of forensic science.” Cardozo Law Review 13: 361
Senese, J. D. 1997. Applied Research Methods in Criminal Justice. Chicago: Nelson-Hall
Publishers.
Shapiro, N.P. and S. Penrod. 1986. “Meta-analysis of facial identification studies.”
Psychological Bulletin. 100(2): 139-56.
Scheck, B., Neufeld, P., and J. Dwyer. 2000. Actual Innocence: Five Days to Execution, and
Other Dispatches From the Wrongly Convicted. New York: Random House.
Silberman C. 1980. Criminal violence, criminal justice. New York: Random House.
Skolnick, J. and J. Fyfe. 1993. Above the Law: Police and Excessive Use of Force. New York:
The Free Press.
Stack, J. and L. Garbus. 1998. The Farm: Angola U.S.A. (film for A&E Network), NY: Gabriel
Films.
Stinson, V., J. Devenport, B. Cutler, and D. Kravitz. 1996. “How effective is the presence-ofcounsel-
safeguard? Attorney perceptions of suggestiveness and fairness based on lineup
procedures.” Journal of Applied Psychology 81:64-75.
Stinson, V., J. Devenport, B. Cutler, and D. Kravitz. 1997. “How effective is the motion-tosuppress-
safeguard? Judges perceptions of suggestiveness and fairness of biased lineup
procedures.” Journal of Applied Psychology 82:211-20.
Strobel, L. 1998. The Case for Christ: A Journalist’s Personal Investigation of the Evidence for
Jesus. Grand Rapids: Zondervan Publishing.
Taylor, B.G., Davis, R.C. and C.D. Maxwell. 2001. “The Effects of a Group Batterer Treatment
Program: A Randomized Experiment in Brooklyn. Justice Quarterly 18, 1: 171-201
Ten More Rampart Cases Voided. 2000 (January 26). Los Angeles Times.
Test Frees Man Jailed for Rape. 2000 (November 26). Cincinnati Enquirer p. 12.
Thirty L.A. Officers Called to Testify for Grand Jury. 2000 (April 14). Los Angeles Times.
U.S. Department of Justice. 1980. Sourcebook of Criminal Justice Statistics.
220
Wagenaar, W. A. and N. Veefkind. 1992. “Comparison of one-person and many-person lineups:
A warning against unsafe practices. In F. Losel, D. Bender, and T. Bliesner (Eds.)
Psychology of Law International Perspectives (pp. 275-85). Berlin: de Gruyter.
Weikel, D. 1994. When the Prosecutor is Guilty. Los Angeles Times, May 13, 1994.
Walker, C. and S. Keir. 1999. eds. Miscarriages of Justice: A Review of Justice in Error.
London:Blackstone.
Walker, S. 1989. Sense and Nonsense About Crime: A Policy Guide. Pacfic Grove,Calf.:
Brooks/Cole
Wells, G.L. 1978. “Applied eyewitness testimony research: System variables and estimator
variables.” Journal of Personality and Social Psychology, 36: 1546-57.
Wells, G.L. 1984. “How adequate is human intuition for judging eyewitness testimony?” In G.L.
Wells and E.F. Loftus (Eds.), Eyewitness Testimony: Psychological Perspectives. New York:
Cambridge University Press.
Wells, G.L. and A.L. Bradfield. 1998. “Good, you identified the suspect. Feedback to
eyewitnesses distorts their reports of the witnessing experience. Journal of Applied
Psychology 83:360-76.
Wells, G.L. and D. Murray. 1983. “What can psychology say about the Neil vs. Biggers criteria
for judging eyewitness identification accuracy? Journal of Applied Psychology, 68: 347-
62.
Wells, G.F. and E. Seelau. 1995. “Eyewitness identification: Psychological research and legal
policy on lineups.” Psychology, Public Policy, and Law 1:765-91.
Wells, G.F., Small, M., Penrod, S., Malpass, R.S., Fulero, S.M., and C.A.E. Brimacombe. 1998.
“Eyewitness Identification Procedures: Recommendations for Lineups and
Photospreads.” Law and Human Behavior, 22, 6: 603-47.
Wilson, J.Q. 1975. Thinking About Crime. New York: Random House.
Wogalter, M.S., Malpass, R.S., and M.A. Burger. 1993. “How police officers conduct line-ups:
A National Survey.” Proceedings of the Human Factors and Ergonomics Society, 37:
640-44.
Yant, M. (1991). Presumed Guilty: When innocent people are wrongly convicted. Buffalo, NY:
Prometheus.
Yarmey, A.D., Yarmey, M.J. and A.L. Yarmey. 1996. “Accuracy of eyewitness identifications in
showups and lineups.” Law and Human Behavior 20:459-77.
Zulawski, D. and D. Wickerland. 1993. Practical Aspects of Interview and Interrogation.
Cases
Ake v. Oklahoma 470 US 68 (1985)
Adams v. State 577 S.W. 2nd. 717 (1979)
Argersinger v. Hamlin 407 U.S. 25 (1972)
Arizona v. Fulminante 499 U.S. 279 (1991)
Berger v. U. S. (1935)
Blevins v. State 108 Ga. App. 738, 134 S.E. 2nd (1963)
Brady v. United States 397 U.SD. 742 (1970)
Commonwealth of Massachusetts v. Cornich, 28 Mass. App. Ct. 173 (1989)
Commonwealth of Virginia v. Walter T. Snyder No. F8118 (1995)
221
Daubert v. Merrell Dow Pharmaceuticals, Inc. 509 U.S. 579 (1993)
Escobedo v. Illinois 378 U.S. 478 (1964)
Folkes v. Chadd 99 Eng.p. 589, 590 (1782).
Frye v. United States 293 F. 1013 (D.C. Cir. 1923)
Gideon v. Wainwright 372 U.S. 335 (1963)
Gilbert v. California 388 U.S. 263 (1967)
Jackson v. Fogg 589 F.2d 108 (2d. Cir.) (1978)
Manson v. Braithwaite 432 U.S. 98 (1977)
Miller v. Pate 386 U.S. 1 (1967)
Miranda v. Arizona 384 U.S. 436 (1966)
Neil v. Biggers 409 U.S. 188 (1972)
North Carolina v. Alford 400 U.S. 25 (1970)
Oklahoma v. Robert Lee Miller CRF-87-963 (1987)
People v. Barad 362 Ill. 584, 200 N.E. 858 (1936)
People v. Harris 28 Cal. 3d 935, 623 P.2d 240 (1981)
People v. Kind 357 Ill. 133, 191 N.E. 244 (1934)
Powell v. Alabama 287 U.S. 45 (1932)b
Rock v. Arkansas 483 U.S. 44 (1987)
Simmons v. United States 390 US 385
State v. Brown 372 A.2nd. 557 (1977)
State v. Cowper 13 Howell’s State Trials 1105 (1699)
State v. Engel 289 N.W. 2d 204 (1980)
State v. Landeros 20 N.J. 76, 118 A.2d 254 (1955)
State of Georgia v. Calvin Crawford Johnson, Jr. CA No. 83 Cr. 12022011-3 (1983)
State of West Virginia v. Harris W.Va. Cir Ct. No. 86-F-442 (1987)
Stovall v. Denno 388 U.S. 293 (1967)
Simmons v. U.S. 390 U.S. 385 (1968)
Timothy Edward Durham v. State of Oklahoma F-93-913 (1993)
U.S. v. Ash 413 U.S. 300 (1973)
U. S. v. Garsson 291 F. 646, 649 (S.D. N.Y. 1923).
U.S. v. Wade 388 U.S. 218 L ed. 2d. (1967)
222
APPENDIX A
EXHIBIT 1
PRE-NOTICE LETTER
UC
University of Cincinnati Division of Criminal Justice
A few days from now you will receive in the mail a request to fill out a brief questionnaire
for an important research project being conducted at the University of Cincinnati. The
questionnaire concerns the perceptions of criminal justice professionals regarding the
frequency and major causes of wrongful conviction (the conviction of innocent people).
Information obtained from people like yourself will be used to assist criminal justice
agencies in better serving the administration of justice.
I am writing you in advance because we have found many people like to know ahead of
time that they will be contacted. It’s only with the generous help of people like you that
our research can be successful.
Thank you for your time and consideration
Dr. James Frank/Robert J. Ramsey
Project Director
223
EXHIBIT 2
INITIAL COVER-LETTER SENT WITH SURVEY
I am writing to ask your assistance in a very important research project which is being conducted
at the University of Cincinnati as part of a dissertation project. The topic of the investigation is
“wrongful conviction,” i.e., the conviction of innocent individuals. Primarily due to
advancements in DNA technology, there have recently been a large number of reported cases of
wrongful conviction around the United States. While it is believed the great majority of
individuals convicted in our criminal courts are in fact guilty, cases involving the conviction of
innocent persons are of concern to many involved in the criminal justice process. This study is
aimed at obtaining a better understanding of the extent and dynamics of wrongful conviction in
order to assist criminal justice agencies in better serving the administration of justice.
Enclosed is a brief questionnaire that is designed to collect valuable information from
knowledgeable persons such as yourself whose position brings them into contact with criminal cases.
The questionnaire takes approximately 10 minutes to complete. I respectfully request your help in
our efforts to collect information on this topic. Your response will be treated with absolute
confidence; our project is concerned with overall patterns of responses, and no individual or his/her
response will ever be identified. You will note that a code number is at the bottom of the last page
of the questionnaire. This is only to allow us to know which questionnaires are missing, so that we
can send a reminder later, where indicated. Once your questionnaire has been received, the code
number will be cut off and destroyed, leaving no way of identifying individual respondents.
If you have any questions about the questionnaire, or the study in general, please contact Robert
Ramsey, Project Director, or Dr. James Frank at any time. When you have completed the
questionnaire, simply put it in the enclosed self-addressed envelope and return it to the
University of Cincinnati. We are estimating that this study will be completed by June 1, 2002.
By completing this questionnaire, you are indicating your consent to participate in the study. If
you would like a summary of the findings when the study is completed, please write your name
and address on the back of the return envelope (not the questionnaire). Thank you in advance for
your assistance in our efforts to better understand the extent and dynamics of wrongful
conviction.
Sincerely,
Robert J. Ramsey, Project Director
James Frank, PHD
University of Cincinnati
Division of Criminal Justice
Phone: (513) 556-5832
224
EXHIBIT 3
FOLLOW-UP “THANK YOU/REMINDER” POSTCARD
UC
University of Cincinnati Division of Criminal Justice
REMINDER !!!!!!
Recently a questionnaire seeking your opinion about the topic of wrongful conviction
was mailed to you. If you have already completed and returned the questionnaire to us,
please accept our sincere thanks. If not, would you please do so today? We are especially
grateful for your help because it is only by asking people like you to share information
about your profession, that we can understand the major causes of wrongful conviction
and its frequency of occurrence.
If you did not receive a questionnaire, or if it was misplaced, please contact Dr. James Frank,
University of Cincinnati Division of Criminal Justice, at 513-556-5832, or e-mail
[email protected], and we will get another one in the mail to you today.
Thank you for your help.
Dr. James Frank/Robert J. Ramsey
Project Directors
225
EXHIBIT 4
COVER LETTER SENT WITH FIRST REPLACEMENT SURVEY
About three weeks ago we sent a questionnaire to you that asked about your perceptions
regarding the wrongful felony conviction of innocent defendants. To the best of our knowledge,
it’s not yet been returned. The comments of people who have already responded include a wide
variety of reasons for the occurrence of wrongful conviction and its perceived frequency. We
think that the results are going to be very useful to criminal justice agencies and members of the
criminal justice profession.
We are writing again because of the importance that your questionnaire has for helping to get
accurate results. Although we sent questionnaires to criminal justice professionals like yourself
in every county in the state, it is only by hearing from nearly everyone in the sample that we can
be sure that the results are truly representative.
A few people have written to say that they no longer are in the criminal justice profession. If this
concern applies to you, we will still like to have you respond to the questionnaire. If you do not
wish to answer the questions, please let us know on the cover of the questionnaire and return it in
the enclosed envelope so that we can delete your name from the mailing list.
A comment on our survey procedures. A questionnaire identification is printed on the back of
the cover of the questionnaire so that we can check you name off the mailing list when it is
returned. The list of names is then destroyed so that individual names can never be connected to
the results in any way. Protecting the confidentially of people’s answers is very important to us,
as well as the University of Cincinnati.
We hope you will fill out and return the questionnaire soon, but if for any reason you prefer not
to answer it, please let us know by returning a note or blank questionnaire in the enclosed
stamped envelope.
Sincerely,
Robert J. Ramsey, Project Director
James Frank, P.H.D.
University of Cincinnati
Department of Criminal Justice
P.O. Box 210389
Cincinnati, Ohio 45221-0389
226
EXHIBIT 5
FINAL COVER LETTER SENT WITH SECOND REPLACEMENT
SURVEY
During the last two months we have sent you several mailings about an important research
project we are conducting at the University of Cincinnati. The purpose of the study is to help
state criminal justice agencies understand the reasons for wrongful conviction, and to assist them
in reducing the frequently of such occurrences.
The study is drawing to a close, and this is the last contact that will be made with the sample of
people who we think, based on their position within the criminal justice system, are best able to
provide the necessary answers to reduce the incidence of wrongful conviction. We are sending
you this final contact by priority mail because of our concern that people who have not
responded may have had different experiences than those who have. Hearing from everyone in
the statewide sample helps assure that the results are as accurate as possible.
Finally, we appreciate your willingness to consider this final request to fill out the questionnaire
in order to give us a better understanding of the problem of wrongful conviction. Your answers
will be held in strict confidentiality; our project is concerned with overall patterns of responses,
and no individual or his/her response will ever be identified. You will note that a code number is
at the top of the first page of the questionnaire. This is only to allow us to know which
questionnaires are missing. Once your questionnaire has been received, the code number will be
cut off and destroyed, leaving no way of identifying individual respondents. Thank you very
much for your time and consideration.
Sincerely,
Robert J. Ramsey, Project Director
James Frank, P.H.D.
University of Cincinnati
Department of Criminal Justice
P.O. Box 210389
Cincinnati, Ohio 45221-0389
P.S. If you do not wish to participate in the study it would be helpful if you would return the
blank questionnaire with a note that you do not wish to participate so we can remove your name
from the mailing list.
227
APPENDIX B
EXHIBIT 6 (A)
SURVEY INSTRUMENT – FRONT COVER
Wrongful Conviction
Survey
When Innocent People Are Convicted of Crimes
A Study of Wrongful Conviction –
Its Extent and Associated Factors.
A Survey of Criminal Justice Professionals
Sponsored by
The University of Cincinnati
Division of Criminal Justice
Please fold and place your completed survey in the stamped self-addressed
envelope and return it to: Division of Criminal Justice, University of
Cincinnati, P.O. Box 210389, Cincinnati, Ohio 45221-0389
228
EXHIBIT 6 (B)
SURVEY INSTRUMENT – INSIDE FRONT COVER
What Is Wrongful Conviction?
Directions. For purposes of this survey the term ‘wrongful conviction’
refers to “people who have been convicted of a criminal offense, but
are in fact innocent.” Most of the questions are multiple-choice;
however, at the end of the survey there is additional space provided
where you may expand upon any response you feel needs additional
comments. It should take between 8 and 10 minutes to complete this
questionnaire.
229
EXHIBIT 6 (C)
SURVEY INSTRUMENT - PAGE 1
WRONGFUL CONVICTION SURVEY
1. Error by eyewitnesses have been found to be associated with some cases of wrongful
conviction. How frequently do you believe eyewitnesses, in good faith, misidentify a
defendant? (Circle One) ↓
a. never b. seldom c. often d. very often
2. How frequently do you believe eyewitnesses intentionally misidentify a defendant? (Circle
One) ↓
a. never b. seldom c. often d. very often
3. In general, how reliable do you believe are the testimony of eyewitnesses in criminal cases?
(Circle One ) ↓
a. very reliable b. usually reliable c. usually unreliable d. very unreliable
4. Error by forensic experts have been found to be associated with some cases of wrongful
conviction. Based on your knowledge and experience, please estimate the frequency of each
error by circling the appropriate number (1 through 9) on the scales below.
Error Made in Good Faith by Forensic Expert ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Intentional Misrepresentation by Forensic Expert ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
5. In general, how reliable do you believe are the scientific conclusions of forensic experts?
(Circle One) ↓
a. very reliable b. usually reliable c. usually unreliable d. very unreliable
230
EXHIBIT 6 (D)
SURVEY INSTRUMENT - PAGE 2
6. Analysis of proven cases of wrongful conviction have indicated five primary
categories of police error. We want to know your opinion concerning how
often, if ever, each of these types of error occurs. Based on your knowledge
and experience, please estimate the frequency of each error by circling the
appropriate number (1 through 9) on the scales below.
Inadequate Police Investigation ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Police Coaching Witnesses In Pretrial I.D. Procedures ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Police Suppressing Exculpatory Evidence ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Police Using False Evidence ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Police Using Undue Pressure To Obtain A Confession ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
7. In general, how reliable do you believe is the evidence presented in court by
police officials? (Circle One) ↓
a. very reliable b. usually reliable c. usually unreliable d. very unreliable
231
EXHIBIT 6 (E)
SURVEY INSTRUMENT - PAGE 3
8. Analysis of proven cases of wrongful conviction have indicated five primary categories of
prosecutorial error. We want to know your opinion concerning how often, if ever, each of
these types of error occur. Based on your knowledge and experience, please estimate the
frequency of each error by circling the appropriate number (1 through 9) on the scales below.
Inadequate Investigation of Case by Prosecutor ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Prosecutor Suppressing Exculpatory Evidence ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Prosecutor Using Undue Plea Bargain Pressure ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Prosecutor Prompting A Witness ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Prosecutor Knowingly Using False Testimony ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
9. In general, how reliable do you believe is the evidence presented in court by prosecutors? ↓
a. very reliable b. usually reliable c. usually unreliable d. very unreliable
232
EXHIBIT 6 (F)
SURVEY INSTRUMENT - PAGE 4
10. Analysis of proven cases of wrongful conviction have indicated five primary categories of
defense attorney error. We want to know your opinion concerning how often, if ever, each of
these types of error occur. Based on your knowledge and experience, please estimate the
frequency of each error by circling the appropriate number (1 through 9) on the scales
below.
Inadequate Investigation of Case by Defense Attorney ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Defense Attorney Failing To File Proper Motions ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Defense Attorney Not Adequately Challenging Forensic Evidence ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Defense Attorney Not Adequately Challenging Witnesses ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Defense Attorney Making Unwarranted Plea Bargain Concessions ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
11. In general, how well do you believe that defense attorneys defend their clients?
a. very well b. moderately well c. not very well d. poorly
233
EXHIBIT 6 (G)
SURVEY INSTRUMENT - PAGE 5
12. Analysis of proven cases of wrongful conviction have indicated four primary categories of
judicial error. We want to know your opinion concerning how often, if ever, each of these
types of error occur. Based on your knowledge and experience, please estimate the
frequency of each error by circling the appropriate number (1 through 9) on the scales below.
Judicial Error Concerning Admissibility Of Physical Evidence ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Judicial Error Concerning Admissibility Of Eyewitness Testimony ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Judicial Error Concerning Admissibility Of Expert Testimony ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
Error Resulting From Judicial Bias ↓
1----------2-----------3-----------4-----------5-----------6-----------7-----------8-----------9
Never Infrequent Moderately Very Always
Frequent Frequent
13. In general, how reliable do you believe are the decisions made by judges concerning
defendants’ guilt or innocence? (Circle One) ↓
a. very reliable b. usually reliable c. moderately reliable d. unreliable
234
EXHIBIT 6 (H)
SURVEY INSTRUMENT - PAGE 6
14. Preliminary analyses of known cases of wrongful conviction suggest that error
can occur at all stages of the criminal justice process. Based on your knowledge
and experience, please rank-order the following (randomly listed) four categories
according to their frequency of errors. (1 = most frequent; 4 = least frequent). ↓
a. errors by defense attorneys . . #_____________
errors by prosecutors . . . . . . #_____________
errors by police officials . . . . #_____________
errors by judges . . . . . . . . . . . #_____________
15. Defendants who are charged with a crime are guilty in: (Circle One) ↓
a. 99-100 % of cases e. 70-79% of cases
b. 95-98 % of cases f. 60-69% of cases
c. 90-94 % of cases g. 50-59 % of cases
d. 80-89 % of cases h. less than 50 % of cases
16. False accusations occur: (Circle One) ↓
a. frequently b. occasionally c. seldom
17. Do you personally know of a case in which an innocent individual confessed to
a crime he or she did not commit? (Circle one) ↓
a. Yes b. No
18. Do you know of a case in your jurisdiction where community pressure to
quickly solve a crime lead to a wrongful conviction? (Circle one) ↓
a. Yes b. No
19. Errors in police identification procedures (e.g. showups, photospreads, lineups,)
have been found to be associated with some cases of wrongful conviction. In
general, how often do you believe that police identification procedures
contribute to misidentification? (Circle One) ↓
a. never b. seldom c. often d. very often
EXHIBIT 6 (I)
SURVEY INSTRUMENT - PAGE 7
20. I would estimate that wrongful conviction occurs in my jurisdiction in _________
235
percent of all felony convictions. (Circle one) ↓
a. 0 % f. 6 % – 10 %
b. less than ½ % g. 11 % – 15 %
c. ½ % - 1 % h. 16 % - 20 %
d. 1 % - 3 % i. 21 % - 25 %
e. 4 % - 5 % j. over 25 %
21. I would estimate that wrongful conviction occurs in the United States in _________
percent of all felony convictions. (Circle one) ↓
a. 0 % f. 6 % – 10 %
b. less than ½ % g. 11 % – 15 %
c. ½ % - 1 % h. 16 % - 20 %
d. 1 % - 3 % i. 21 % - 25 %
e. 4 % - 5 % j. over 25 %
22. What would you consider an acceptable level of inevitable wrongful convictions?
(Circle One) ↓
a. 0 % f. 6 % – 10 %
b. less than ½ % g. 11 % – 15 %
c. ½ % - 1 % h. 16 % - 20 %
d. 1 % - 3 % i. 21 % - 25 %
e. 4 % - 5 % j. over 25 %
23. Do you support the use of the death penalty in the United States? (Circle One) ↓
a. Yes b. No
24. In your opinion, do wrongful convictions occur frequently enough to warrant
procedural changes in the criminal justice system? (Circle One) ↓
a. Yes b. No
236
EXHIBIT 6 (J)
SURVEY INSTRUMENT - PAGE 8
25. Have you had experience as a police officer? …. YES NO # of years ______
“ “ “ “ “ “ prosecutor? ………YES NO # of years ______
“ “ “ “ private defense lawyer? YES NO # of years ______
“ “ “ “ “ “ public defender? …….YES NO # of years _____
“ “ “ “ “ “ judge? …………… YES NO # of years ______
26. What is your present jurisdiction? (Circle one or more) ↓
a. Village b. Township c. City d. County e. State f. Other_________
(please specify)
27. What is your gender? (Circle One) ↓
a. Male b. Female
28. In which of the following racial categories do you consider yourself? (Circle One) ↓
a. White b. Hispanic c. Black d. Asian e. Other ________________
(please specify)
29. What is your age? (Circle One) ↓
a. 18-29 b. 30-39 c. 40-49 d. 50-59 e. 60-69 f. 70-79 g. 80+
30. In your opinion, what steps could be taken by criminal justice professionals to
reduce the incidence of wrongful conviction?
___________________________________________________________________
___________________________________________________________________
____________________________________________________________________
____________________________________________________________________
237
EXHIBIT 6 (K)
INSIDE BACK COVER
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
238
EXHIBIT 6 (L)
SURVEY INSTRUMENT – BACK COVER
You Are Finished.
THANK YOU FOR PARTICIPATING
IN THIS IMPORTANT STUDY !
Your contribution to this project is greatly appreciated.
If you would like a summary of results, please print your
name and address on the back of the return envelope
(NOT the questionnaire), and you will be mailed a copy.
Please fold and place your completed survey in the stamped self-addressed
envelope and return it to: Division of Criminal Justice, University of
Cincinnati, P.O. Box 210389, Cincinnati, Ohio 45221-0389